Difference between revisions of "Neural Networks"

From Chessprogramming wiki
Jump to: navigation, search
(35 intermediate revisions by the same user not shown)
Line 18: Line 18:
 
The [https://en.wikipedia.org/wiki/Perceptron perceptron] is an algorithm for [[Supervised Learning|supervised learning]] of [https://en.wikipedia.org/wiki/Binary_classification binary classifiers]. It was the first artificial neural network, introduced in 1957 by [https://en.wikipedia.org/wiki/Frank_Rosenblatt Frank Rosenblatt] <ref>[https://en.wikipedia.org/wiki/Frank_Rosenblatt Frank Rosenblatt] ('''1957'''). ''The Perceptron - a Perceiving and Recognizing Automaton''. Report 85-460-1, [https://en.wikipedia.org/wiki/Calspan#History Cornell Aeronautical Laboratory]</ref>, implemented in custom hardware. In its basic form it consists of a single neuron with multiple inputs and associated weights.
 
The [https://en.wikipedia.org/wiki/Perceptron perceptron] is an algorithm for [[Supervised Learning|supervised learning]] of [https://en.wikipedia.org/wiki/Binary_classification binary classifiers]. It was the first artificial neural network, introduced in 1957 by [https://en.wikipedia.org/wiki/Frank_Rosenblatt Frank Rosenblatt] <ref>[https://en.wikipedia.org/wiki/Frank_Rosenblatt Frank Rosenblatt] ('''1957'''). ''The Perceptron - a Perceiving and Recognizing Automaton''. Report 85-460-1, [https://en.wikipedia.org/wiki/Calspan#History Cornell Aeronautical Laboratory]</ref>, implemented in custom hardware. In its basic form it consists of a single neuron with multiple inputs and associated weights.
  
[[Supervised learning]] is applied using a set D of labeled [https://en.wikipedia.org/wiki/Test_set training data] with pairs of [https://en.wikipedia.org/wiki/Feature_vector feature vectors] (x) and given results as desired output (d), usually started with cleared or randomly initialized weight vector w. The output is calculated by all inputs of a sample, multiplied by its corresponding weights, passing the sum to the activation function f. The difference of desired and actual value is then immediately used modify the weights for all features using a learning rate 0.0 < α <= 1.0:
+
[[Supervised Learning|Supervised learning]] is applied using a set D of labeled [https://en.wikipedia.org/wiki/Test_set training data] with pairs of [https://en.wikipedia.org/wiki/Feature_vector feature vectors] (x) and given results as desired output (d), usually started with cleared or randomly initialized weight vector w. The output is calculated by all inputs of a sample, multiplied by its corresponding weights, passing the sum to the activation function f. The difference of desired and actual value is then immediately used modify the weights for all features using a learning rate 0.0 < α <= 1.0:
 
<pre>
 
<pre>
 
   for (j=0, Σ = 0.0; j < nSamples; ++j) {
 
   for (j=0, Σ = 0.0; j < nSamples; ++j) {
Line 35: Line 35:
  
 
==Backpropagation==
 
==Backpropagation==
In 1974, [https://en.wikipedia.org/wiki/Paul_Werbos Paul Werbos] started to end the AI winter concerning neural networks, when he first described the mathematical process of training [https://en.wikipedia.org/wiki/Multilayer_perceptron multilayer perceptrons] through [https://en.wikipedia.org/wiki/Backpropagation backpropagation] of errors <ref>[https://en.wikipedia.org/wiki/Paul_Werbos Paul Werbos] ('''1974'''). ''[http://aitopics.org/publication/beyond-regression-new-tools-prediction-and-analysis-behavioral-sciences Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences]''. Ph. D. thesis, [[Harvard University]]</ref>, derived in the context of [https://en.wikipedia.org/wiki/Control_theory control theory] by [https://en.wikipedia.org/wiki/Henry_J._Kelley Henry J. Kelley] in 1960 <ref>[https://en.wikipedia.org/wiki/Henry_J._Kelley Henry J. Kelley] ('''1960'''). ''[http://arc.aiaa.org/doi/abs/10.2514/8.5282?journalCode=arsj& Gradient Theory of Optimal Flight Paths]''. [http://arc.aiaa.org/loi/arsj ARS Journal, Vol. 30, No. 10</ref> and by [https://en.wikipedia.org/wiki/Arthur_E._Bryson Arthur E. Bryson] in 1961 <ref>[https://en.wikipedia.org/wiki/Arthur_E._Bryson Arthur E. Bryson] ('''1961'''). ''A gradient method for optimizing multi-stage allocation processes''. In Proceedings of the [[Harvard University]] Symposium on digital computers and their applications</ref> using principles of [[Dynamic Programming|dynamic programming]], simplified by [https://en.wikipedia.org/wiki/Stuart_Dreyfus Stuart Dreyfus] in 1961 applying the [https://en.wikipedia.org/wiki/Chain_rule chain rule] <ref>[https://en.wikipedia.org/wiki/Stuart_Dreyfus Stuart Dreyfus] ('''1961'''). ''[http://www.rand.org/pubs/papers/P2374.html The numerical solution of variational problems]''. RAND paper P-2374</ref>. It was in 1982, when Werbos applied a [https://en.wikipedia.org/wiki/Automatic_differentiation automatic differentiation] method described in 1970 by [[Mathematician#SLinnainmaa|Seppo Linnainmaa]] <ref>[[Mathematician#SLinnainmaa|Seppo Linnainmaa]] ('''1970'''). ''The representation of the cumulative rounding error of an algorithm as a Taylor expansion of the local rounding errors''. Master's thesis, [https://en.wikipedia.org/wiki/University_of_Helsinki University of Helsinki]</ref> to neural networks in the way that is widely used today <ref>[https://en.wikipedia.org/wiki/Paul_Werbos Paul Werbos] ('''1982'''). ''Applications of advances in nonlinear sensitivity analysis''. [http://link.springer.com/book/10.1007%2FBFb0006119 System Modeling and Optimization], [https://en.wikipedia.org/wiki/Springer_Science%2BBusiness_Media Springer], [http://werbos.com/Neural/SensitivityIFIPSeptember1981.pdf pdf]</ref> <ref>[https://en.wikipedia.org/wiki/Paul_Werbos Paul Werbos] ('''1994'''). ''[http://eu.wiley.com/WileyCDA/WileyTitle/productCd-0471598976.html The Roots of Backpropagation. From Ordered Derivatives to Neural Networks and Political Forecasting]''. [https://en.wikipedia.org/wiki/John_Wiley_%26_Sons John Wiley & Sons]</ref> <ref>[http://www.scholarpedia.org/article/Deep_Learning#Backpropagation Deep Learning - Scholarpedia | Backpropagation] by [[Jürgen Schmidhuber]]</ref> <ref>[http://people.idsia.ch/~juergen/who-invented-backpropagation.html Who Invented Backpropagation?]  by [[Jürgen Schmidhuber]] (2014, 2015)</ref>.  
+
In 1974, [[Mathematician#PWerbos|Paul Werbos]] started to end the AI winter concerning neural networks, when he first described the mathematical process of training [https://en.wikipedia.org/wiki/Multilayer_perceptron multilayer perceptrons] through [https://en.wikipedia.org/wiki/Backpropagation backpropagation] of errors <ref>[[Mathematician#PWerbos|Paul Werbos]] ('''1974'''). ''Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences''. Ph. D. thesis, [[Harvard University]]</ref>, derived in the context of [https://en.wikipedia.org/wiki/Control_theory control theory] by [https://en.wikipedia.org/wiki/Henry_J._Kelley Henry J. Kelley] in 1960 <ref>[https://en.wikipedia.org/wiki/Henry_J._Kelley Henry J. Kelley] ('''1960'''). ''[http://arc.aiaa.org/doi/abs/10.2514/8.5282?journalCode=arsj& Gradient Theory of Optimal Flight Paths]''. [http://arc.aiaa.org/loi/arsj ARS Journal, Vol. 30, No. 10</ref> and by [https://en.wikipedia.org/wiki/Arthur_E._Bryson Arthur E. Bryson] in 1961 <ref>[https://en.wikipedia.org/wiki/Arthur_E._Bryson Arthur E. Bryson] ('''1961'''). ''A gradient method for optimizing multi-stage allocation processes''. In Proceedings of the [[Harvard University]] Symposium on digital computers and their applications</ref> using principles of [[Dynamic Programming|dynamic programming]], simplified by [[Mathematician#SEDreyfus|Stuart E. Dreyfus]] in 1961 applying the [https://en.wikipedia.org/wiki/Chain_rule chain rule] <ref>[[Mathematician#SEDreyfus|Stuart E. Dreyfus]] ('''1961'''). ''[http://www.rand.org/pubs/papers/P2374.html The numerical solution of variational problems]''. RAND paper P-2374</ref>. It was in 1982, when Werbos applied a [https://en.wikipedia.org/wiki/Automatic_differentiation automatic differentiation] method described in 1970 by [[Mathematician#SLinnainmaa|Seppo Linnainmaa]] <ref>[[Mathematician#SLinnainmaa|Seppo Linnainmaa]] ('''1970'''). ''The representation of the cumulative rounding error of an algorithm as a Taylor expansion of the local rounding errors''. Master's thesis, [https://en.wikipedia.org/wiki/University_of_Helsinki University of Helsinki]</ref> to neural networks in the way that is widely used today <ref>[[Mathematician#PWerbos|Paul Werbos]] ('''1982'''). ''Applications of advances in nonlinear sensitivity analysis''. [http://link.springer.com/book/10.1007%2FBFb0006119 System Modeling and Optimization], [https://en.wikipedia.org/wiki/Springer_Science%2BBusiness_Media Springer], [http://werbos.com/Neural/SensitivityIFIPSeptember1981.pdf pdf]</ref> <ref>[[Mathematician#PWerbos|Paul Werbos]] ('''1994'''). ''The Roots of Backpropagation. From Ordered Derivatives to Neural Networks and Political Forecasting''. [https://en.wikipedia.org/wiki/John_Wiley_%26_Sons John Wiley & Sons]</ref> <ref>[http://www.scholarpedia.org/article/Deep_Learning#Backpropagation Deep Learning - Scholarpedia | Backpropagation] by [[Jürgen Schmidhuber]]</ref> <ref>[http://people.idsia.ch/~juergen/who-invented-backpropagation.html Who Invented Backpropagation?]  by [[Jürgen Schmidhuber]] (2014, 2015)</ref>.  
  
 
Backpropagation is a generalization of the [https://en.wikipedia.org/wiki/Delta_rule delta] rule to multilayered [https://en.wikipedia.org/wiki/Feedforward_neural_network feedforward networks], made possible by using the [https://en.wikipedia.org/wiki/Chain_rule chain rule] to iteratively compute [https://en.wikipedia.org/wiki/Gradient gradients] for each layer. Backpropagation requires that the [https://en.wikipedia.org/wiki/Activation_function activation function] used by the artificial neurons be [https://en.wikipedia.org/wiki/Differentiable_function differentiable], which is true for the common [https://en.wikipedia.org/wiki/Sigmoid_function sigmoid] [https://en.wikipedia.org/wiki/Logistic_function logistic function] or its [https://en.wikipedia.org/wiki/Softmax_function softmax] generalization in [https://en.wikipedia.org/wiki/Multiclass_classification multiclass classification].  
 
Backpropagation is a generalization of the [https://en.wikipedia.org/wiki/Delta_rule delta] rule to multilayered [https://en.wikipedia.org/wiki/Feedforward_neural_network feedforward networks], made possible by using the [https://en.wikipedia.org/wiki/Chain_rule chain rule] to iteratively compute [https://en.wikipedia.org/wiki/Gradient gradients] for each layer. Backpropagation requires that the [https://en.wikipedia.org/wiki/Activation_function activation function] used by the artificial neurons be [https://en.wikipedia.org/wiki/Differentiable_function differentiable], which is true for the common [https://en.wikipedia.org/wiki/Sigmoid_function sigmoid] [https://en.wikipedia.org/wiki/Logistic_function logistic function] or its [https://en.wikipedia.org/wiki/Softmax_function softmax] generalization in [https://en.wikipedia.org/wiki/Multiclass_classification multiclass classification].  
Line 96: Line 96:
 
===Alpha Zero===
 
===Alpha Zero===
 
In December 2017, the [[Google]] [[DeepMind]] team along with former [[Giraffe]] author [[Matthew Lai]] reported on their generalized [[AlphaZero]] algorithm, combining [[Deep Learning|Deep learning]] with [[Monte-Carlo Tree Search]]. AlphaZero can achieve, tabula rasa, superhuman performance in many challenging domains with some training effort. Starting from random play, and given no domain knowledge except the game rules, AlphaZero achieved a superhuman level of play in the games of chess and [[Shogi]] as well as Go, and convincingly defeated a world-champion program in each case <ref>[[David Silver]], [[Thomas Hubert]], [[Julian Schrittwieser]], [[Ioannis Antonoglou]], [[Matthew Lai]], [[Arthur Guez]], [[Marc Lanctot]], [[Laurent Sifre]], [[Dharshan Kumaran]], [[Thore Graepel]], [[Timothy Lillicrap]], [[Karen Simonyan]], [[Demis Hassabis]] ('''2017'''). ''Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm''. [https://arxiv.org/abs/1712.01815 arXiv:1712.01815]</ref>.  
 
In December 2017, the [[Google]] [[DeepMind]] team along with former [[Giraffe]] author [[Matthew Lai]] reported on their generalized [[AlphaZero]] algorithm, combining [[Deep Learning|Deep learning]] with [[Monte-Carlo Tree Search]]. AlphaZero can achieve, tabula rasa, superhuman performance in many challenging domains with some training effort. Starting from random play, and given no domain knowledge except the game rules, AlphaZero achieved a superhuman level of play in the games of chess and [[Shogi]] as well as Go, and convincingly defeated a world-champion program in each case <ref>[[David Silver]], [[Thomas Hubert]], [[Julian Schrittwieser]], [[Ioannis Antonoglou]], [[Matthew Lai]], [[Arthur Guez]], [[Marc Lanctot]], [[Laurent Sifre]], [[Dharshan Kumaran]], [[Thore Graepel]], [[Timothy Lillicrap]], [[Karen Simonyan]], [[Demis Hassabis]] ('''2017'''). ''Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm''. [https://arxiv.org/abs/1712.01815 arXiv:1712.01815]</ref>.  
 +
<span id="engines"></span>
 +
===NN Chess Programs===
 +
* [[:Category:NN]]
  
 
=See also=
 
=See also=
Line 112: Line 115:
 
* [[Pattern Recognition]]
 
* [[Pattern Recognition]]
 
* [[Temporal Difference Learning]]
 
* [[Temporal Difference Learning]]
<span id="engines"></span>
 
=NN Chess Programs=
 
* [[Alexs]]
 
* [[AlphaZero]]
 
* [[Arminius]]
 
* [[Blondie25]]
 
* [[ChessMaps]]
 
* [[Chessterfield]]
 
* [[Deep Pink]]
 
* [[Giraffe]]
 
* [[Golch]]
 
* [[Gosu]]
 
* [[Hermann]]
 
* [[Leela Chess Zero]]
 
* [[Morph]]
 
* [[NeuroChess]]
 
* [[Octavius]]
 
* [[SAL]]
 
* [[Scorpio]]
 
* [[Spawkfish]]
 
* [[Stoofvlees]]
 
* [[Tempo (engine)|Tempo]]
 
* [[Zurichess]]
 
  
 
=Selected Publications=
 
=Selected Publications=
Line 154: Line 134:
 
* [https://en.wikipedia.org/wiki/Henry_J._Kelley Henry J. Kelley] ('''1960'''). ''[http://arc.aiaa.org/doi/abs/10.2514/8.5282?journalCode=arsj& Gradient Theory of Optimal Flight Paths]''. [http://arc.aiaa.org/loi/arsj ARS Journal, Vol. 30, No. 10 » [[Neural Networks#Backpropagation|Backpropagation]]
 
* [https://en.wikipedia.org/wiki/Henry_J._Kelley Henry J. Kelley] ('''1960'''). ''[http://arc.aiaa.org/doi/abs/10.2514/8.5282?journalCode=arsj& Gradient Theory of Optimal Flight Paths]''. [http://arc.aiaa.org/loi/arsj ARS Journal, Vol. 30, No. 10 » [[Neural Networks#Backpropagation|Backpropagation]]
 
* [https://en.wikipedia.org/wiki/Arthur_E._Bryson Arthur E. Bryson] ('''1961'''). ''A gradient method for optimizing multi-stage allocation processes''. In Proceedings of the [[Harvard University]] Symposium on digital computers and their applications » [[Neural Networks#Backpropagation|Backpropagation]]
 
* [https://en.wikipedia.org/wiki/Arthur_E._Bryson Arthur E. Bryson] ('''1961'''). ''A gradient method for optimizing multi-stage allocation processes''. In Proceedings of the [[Harvard University]] Symposium on digital computers and their applications » [[Neural Networks#Backpropagation|Backpropagation]]
* [https://en.wikipedia.org/wiki/Stuart_Dreyfus Stuart Dreyfus] ('''1961'''). ''[http://www.rand.org/pubs/papers/P2374.html The numerical solution of variational problems]''. RAND paper P-2374 » [[Neural Networks#Backpropagation|Backpropagation]]
+
* [[Mathematician#SEDreyfus|Stuart E. Dreyfus]] ('''1961'''). ''[http://www.rand.org/pubs/papers/P2374.html The numerical solution of variational problems]''. RAND paper P-2374 » [[Neural Networks#Backpropagation|Backpropagation]]
 
* [https://en.wikipedia.org/wiki/Frank_Rosenblatt Frank Rosenblatt] ('''1962'''). ''[http://catalog.hathitrust.org/Record/000203591 Principles of Neurodynamics: Perceptrons and the Theory of Brain Mechanisms]''. Spartan Books
 
* [https://en.wikipedia.org/wiki/Frank_Rosenblatt Frank Rosenblatt] ('''1962'''). ''[http://catalog.hathitrust.org/Record/000203591 Principles of Neurodynamics: Perceptrons and the Theory of Brain Mechanisms]''. Spartan Books
 
* [https://en.wikipedia.org/wiki/Alexey_Grigorevich_Ivakhnenko Alexey G. Ivakhnenko] ('''1965'''). ''Cybernetic Predicting Devices''. [https://en.wikipedia.org/wiki/Naukova_Dumka Naukova Dumka]
 
* [https://en.wikipedia.org/wiki/Alexey_Grigorevich_Ivakhnenko Alexey G. Ivakhnenko] ('''1965'''). ''Cybernetic Predicting Devices''. [https://en.wikipedia.org/wiki/Naukova_Dumka Naukova Dumka]
Line 165: Line 145:
 
* [[Mathematician#SGrossberg|Stephen Grossberg]] ('''1973'''). ''Contour Enhancement, Short Term Memory, and Constancies in Reverberating Neural Networks''. [https://en.wikipedia.org/wiki/Studies_in_Applied_Mathematics Studies in Applied Mathematics], Vol. 52, [http://cns.bu.edu/~steve/Gro1973StudiesAppliedMath.pdf pdf]
 
* [[Mathematician#SGrossberg|Stephen Grossberg]] ('''1973'''). ''Contour Enhancement, Short Term Memory, and Constancies in Reverberating Neural Networks''. [https://en.wikipedia.org/wiki/Studies_in_Applied_Mathematics Studies in Applied Mathematics], Vol. 52, [http://cns.bu.edu/~steve/Gro1973StudiesAppliedMath.pdf pdf]
 
* [[Mathematician#SGrossberg|Stephen Grossberg]] ('''1974'''). ''[http://techlab.bu.edu/resources/article_view/classical_and_instrumental_learning_by_neural_networks/ Classical and instrumental learning by neural networks]''. Progress in Theoretical Biology. [https://en.wikipedia.org/wiki/Academic_Press Academic Press]
 
* [[Mathematician#SGrossberg|Stephen Grossberg]] ('''1974'''). ''[http://techlab.bu.edu/resources/article_view/classical_and_instrumental_learning_by_neural_networks/ Classical and instrumental learning by neural networks]''. Progress in Theoretical Biology. [https://en.wikipedia.org/wiki/Academic_Press Academic Press]
* [https://en.wikipedia.org/wiki/Paul_Werbos Paul Werbos] ('''1974'''). ''[http://aitopics.org/publication/beyond-regression-new-tools-prediction-and-analysis-behavioral-sciences Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences]''. Ph. D. thesis, [[Harvard University]] <ref>[https://en.wikipedia.org/wiki/Backpropagation Backpropagation from Wikipedia]</ref> <ref>[https://en.wikipedia.org/wiki/Paul_Werbos Paul Werbos] ('''1994'''). ''[http://eu.wiley.com/WileyCDA/WileyTitle/productCd-0471598976.html The Roots of Backpropagation. From Ordered Derivatives to Neural Networks and Political Forecasting]''. [https://en.wikipedia.org/wiki/John_Wiley_%26_Sons John Wiley & Sons]</ref>
+
* [[Mathematician#PWerbos|Paul Werbos]] ('''1974'''). ''Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences''. Ph. D. thesis, [[Harvard University]] <ref>[https://en.wikipedia.org/wiki/Backpropagation Backpropagation from Wikipedia]</ref> <ref>[[Mathematician#PWerbos|Paul Werbos]] ('''1994'''). ''The Roots of Backpropagation. From Ordered Derivatives to Neural Networks and Political Forecasting''. [https://en.wikipedia.org/wiki/John_Wiley_%26_Sons John Wiley & Sons]</ref>
 
* [[Richard Sutton]] ('''1978'''). ''Single channel theory: A neuronal theory of learning''. Brain Theory Newsletter 3, No. 3/4, pp. 72-75. [http://www.cs.ualberta.ca/%7Esutton/papers/sutton-78-BTN.pdf pdf]
 
* [[Richard Sutton]] ('''1978'''). ''Single channel theory: A neuronal theory of learning''. Brain Theory Newsletter 3, No. 3/4, pp. 72-75. [http://www.cs.ualberta.ca/%7Esutton/papers/sutton-78-BTN.pdf pdf]
 
==1980 ...==
 
==1980 ...==
 
* [http://www.scholarpedia.org/article/User:Kunihiko_Fukushima Kunihiko Fukushima] ('''1980'''). ''Neocognitron: A Self-organizing Neural Network Model for a Mechanism of Pattern Recognition Unaffected by Shift in Position''. [http://link.springer.com/journal/422 Biological Cybernetics], Vol. 36 <ref>[http://www.scholarpedia.org/article/Neocognitron Neocognitron - Scholarpedia] by [http://www.scholarpedia.org/article/User:Kunihiko_Fukushima Kunihiko Fukushima]</ref>
 
* [http://www.scholarpedia.org/article/User:Kunihiko_Fukushima Kunihiko Fukushima] ('''1980'''). ''Neocognitron: A Self-organizing Neural Network Model for a Mechanism of Pattern Recognition Unaffected by Shift in Position''. [http://link.springer.com/journal/422 Biological Cybernetics], Vol. 36 <ref>[http://www.scholarpedia.org/article/Neocognitron Neocognitron - Scholarpedia] by [http://www.scholarpedia.org/article/User:Kunihiko_Fukushima Kunihiko Fukushima]</ref>
 
* [[Richard Sutton]], [[Andrew Barto]] ('''1981'''). ''Toward a modern theory of adaptive networks: Expectation and prediction''. Psychological Review, Vol. 88, pp. 135-170. [http://www.cs.ualberta.ca/%7Esutton/papers/sutton-barto-81-PsychRev.pdf pdf]
 
* [[Richard Sutton]], [[Andrew Barto]] ('''1981'''). ''Toward a modern theory of adaptive networks: Expectation and prediction''. Psychological Review, Vol. 88, pp. 135-170. [http://www.cs.ualberta.ca/%7Esutton/papers/sutton-barto-81-PsychRev.pdf pdf]
* [https://en.wikipedia.org/wiki/Paul_Werbos Paul Werbos] ('''1982'''). ''Applications of advances in nonlinear sensitivity analysis''. [http://link.springer.com/book/10.1007%2FBFb0006119 System Modeling and Optimization], [https://en.wikipedia.org/wiki/Springer_Science%2BBusiness_Media Springer], [http://werbos.com/Neural/SensitivityIFIPSeptember1981.pdf pdf]
+
* [[Mathematician#PWerbos|Paul Werbos]] ('''1982'''). ''Applications of advances in nonlinear sensitivity analysis''. [http://link.springer.com/book/10.1007%2FBFb0006119 System Modeling and Optimization], [https://en.wikipedia.org/wiki/Springer_Science%2BBusiness_Media Springer], [http://werbos.com/Neural/SensitivityIFIPSeptember1981.pdf pdf]
 
* [[A. Harry Klopf]] ('''1982'''). ''The Hedonistic Neuron: A Theory of Memory, Learning, and Intelligence''. Hemisphere Publishing Corporation, [[University of Michigan]]
 
* [[A. Harry Klopf]] ('''1982'''). ''The Hedonistic Neuron: A Theory of Memory, Learning, and Intelligence''. Hemisphere Publishing Corporation, [[University of Michigan]]
 
* [[Mathematician#DHAckley|David H. Ackley]], [[Mathematician#GEHinton|Geoffrey E. Hinton]], [[Terrence J. Sejnowski]] ('''1985'''). ''A Learning Algorithm for Boltzmann Machines''. Cognitive Science, Vol. 9, No. 1, [https://web.archive.org/web/20110718022336/http://learning.cs.toronto.edu/~hinton/absps/cogscibm.pdf pdf]
 
* [[Mathematician#DHAckley|David H. Ackley]], [[Mathematician#GEHinton|Geoffrey E. Hinton]], [[Terrence J. Sejnowski]] ('''1985'''). ''A Learning Algorithm for Boltzmann Machines''. Cognitive Science, Vol. 9, No. 1, [https://web.archive.org/web/20110718022336/http://learning.cs.toronto.edu/~hinton/absps/cogscibm.pdf pdf]
Line 188: Line 168:
 
* [[Eric B. Baum]] ('''1989'''). ''[http://papers.nips.cc/paper/226-the-perceptron-algorithm-is-fast-for-non-malicious-distributions The Perceptron Algorithm Is Fast for Non-Malicious Distributions]''. [http://papers.nips.cc/book/advances-in-neural-information-processing-systems-2-1989 NIPS 1989]
 
* [[Eric B. Baum]] ('''1989'''). ''[http://papers.nips.cc/paper/226-the-perceptron-algorithm-is-fast-for-non-malicious-distributions The Perceptron Algorithm Is Fast for Non-Malicious Distributions]''. [http://papers.nips.cc/book/advances-in-neural-information-processing-systems-2-1989 NIPS 1989]
 
* [[Eric B. Baum]] ('''1989'''). ''[http://www.mitpressjournals.org/doi/abs/10.1162/neco.1989.1.2.201#.VfGX0JdpluM A Proposal for More Powerful Learning Algorithms]''. [https://en.wikipedia.org/wiki/Neural_Computation_%28journal%29 Neural Computation], Vol. 1, No. 2
 
* [[Eric B. Baum]] ('''1989'''). ''[http://www.mitpressjournals.org/doi/abs/10.1162/neco.1989.1.2.201#.VfGX0JdpluM A Proposal for More Powerful Learning Algorithms]''. [https://en.wikipedia.org/wiki/Neural_Computation_%28journal%29 Neural Computation], Vol. 1, No. 2
* [http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/i/Irani:E=_A=.html Erach A. Irani], [http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/m/Matts:John_P=.html John P. Matts], [http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/l/Long:John_M=.html John M. Long], [[James R. Slagle]], POSCH group ('''1989'''). ''Using Artificial Neural Nets for Statistical Discovery: Observations after Using Backpropogation, Expert Systems, and Multiple-Linear Regression on Clinical Trial Data''. University of Minnesota, Minneapolis, MN 55455, USA, Complex Systems 3, [http://www.complex-systems.com/pdf/03-3-5.pdf pdf]
+
* [http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/i/Irani:E=_A=.html Erach A. Irani], [http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/m/Matts:John_P=.html John P. Matts], [http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/l/Long:John_M=.html John M. Long], [[James R. Slagle]], POSCH group ('''1989'''). ''Using Artificial Neural Nets for Statistical Discovery: Observations after Using Backpropogation, Expert Systems, and Multiple-Linear Regression on Clinical Trial Data''. [[University of Minnesota]], Minneapolis, MN 55455, USA, Complex Systems 3, [http://www.complex-systems.com/pdf/03-3-5.pdf pdf]
 
* [[Gerald Tesauro]], [[Terrence J. Sejnowski]] ('''1989'''). ''A Parallel Network that Learns to Play Backgammon''. [https://en.wikipedia.org/wiki/Artificial_Intelligence_%28journal%29 Artificial Intelligence], Vol. 39, No. 3
 
* [[Gerald Tesauro]], [[Terrence J. Sejnowski]] ('''1989'''). ''A Parallel Network that Learns to Play Backgammon''. [https://en.wikipedia.org/wiki/Artificial_Intelligence_%28journal%29 Artificial Intelligence], Vol. 39, No. 3
 
* [[Mathematician#EGelenbe|Erol Gelenbe]] ('''1989'''). ''[http://cognet.mit.edu/journal/10.1162/neco.1989.1.4.502 Random Neural Networks with Negative and Positive Signals and Product Form Solution]''. [https://en.wikipedia.org/wiki/Neural_Computation_(journal) Neural Computation], Vol. 1, No. 4
 
* [[Mathematician#EGelenbe|Erol Gelenbe]] ('''1989'''). ''[http://cognet.mit.edu/journal/10.1162/neco.1989.1.4.502 Random Neural Networks with Negative and Positive Signals and Product Form Solution]''. [https://en.wikipedia.org/wiki/Neural_Computation_(journal) Neural Computation], Vol. 1, No. 4
 +
* [[Mathematician#XZhang|Xiru Zhang]], [https://dblp.uni-trier.de/pers/hd/m/McKenna:Michael Michael McKenna], [[Mathematician#JPMesirov|Jill P. Mesirov]], [[David Waltz]] ('''1989'''). ''[http://papers.neurips.cc/paper/281-an-efficient-implementation-of-the-back-propagation-algorithm-on-the-connection-machine-cm-2 An Efficient Implementation of the Back-propagation Algorithm on the Connection Machine CM-2]''. [https://dblp.uni-trier.de/db/conf/nips/nips1989.html NIPS 1989]
 
==1990 ...==
 
==1990 ...==
* [https://en.wikipedia.org/wiki/Paul_Werbos Paul Werbos] ('''1990'''). ''Backpropagation Through Time: What It Does and How to Do It''. Proceedings of the [[IEEE]], Vol. 78, No. 10, [http://deeplearning.cs.cmu.edu/pdfs/Werbos.backprop.pdf pdf]
+
* [[Mathematician#PWerbos|Paul Werbos]] ('''1990'''). ''Backpropagation Through Time: What It Does and How to Do It''. Proceedings of the [[IEEE]], Vol. 78, No. 10, [http://deeplearning.cs.cmu.edu/pdfs/Werbos.backprop.pdf pdf]
 
* [[Gordon Goetsch]] ('''1990'''). ''Maximization of Mutual Information in a Context Sensitive Neural Network''. Ph.D. thesis
 
* [[Gordon Goetsch]] ('''1990'''). ''Maximization of Mutual Information in a Context Sensitive Neural Network''. Ph.D. thesis
 
* [[Vadim Anshelevich]] ('''1990'''). ''Neural Networks''. Review. in Multi Component Systems (Russian)
 
* [[Vadim Anshelevich]] ('''1990'''). ''Neural Networks''. Review. in Multi Component Systems (Russian)
 
* [[Eric B. Baum]] ('''1990'''). ''Polynomial Time Algorithms for Learning Neural Nets''. [http://dblp.uni-trier.de/db/conf/colt/colt1990.html#Baum90 COLT 1990]
 
* [[Eric B. Baum]] ('''1990'''). ''Polynomial Time Algorithms for Learning Neural Nets''. [http://dblp.uni-trier.de/db/conf/colt/colt1990.html#Baum90 COLT 1990]
 +
* [https://dblp.uni-trier.de/pers/hd/r/Ruck:Dennis_W= Dennis W. Ruck], [http://spie.org/profile/Steven.Rogers-5480?SSO=1 Steven K. Rogers], [https://dblp.uni-trier.de/pers/hd/k/Kabrisky:Matthew Matthew Kabrisky], [[Mathematician#MEOxley|Mark E. Oxley]], [[Bruce W. Suter]] ('''1990'''). ''[https://ieeexplore.ieee.org/document/80266 The multilayer perceptron as an approximation to a Bayes optimal discriminant function]''.  [[IEEE#NN|IEEE Transactions on Neural Networks]], Vol. 1, No. 4
 +
* [https://dblp.uni-trier.de/pers/hd/h/Hellstrom:Benjamin_J= Benjamin J. Hellstrom], [[Laveen Kanal|Laveen N. Kanal]] ('''1990'''). ''[https://ieeexplore.ieee.org/document/5726889 The definition of necessary hidden units in neural networks for combinatorial optimization]''. [https://dblp.uni-trier.de/db/conf/ijcnn/ijcnn1990.html IJCNN 1990]
 +
* [[Mathematician#XZhang|Xiru Zhang]], [https://dblp.uni-trier.de/pers/hd/m/McKenna:Michael Michael McKenna], [[Mathematician#JPMesirov|Jill P. Mesirov]], [[David Waltz]] ('''1990'''). ''[https://www.sciencedirect.com/science/article/pii/016781919090084M The backpropagation algorithm on grid and hypercube architectures]''. [https://www.journals.elsevier.com/parallel-computing Parallel Computing], Vol. 14, No. 3
 
'''1991'''
 
'''1991'''
 
* [[Mathematician#SHochreiter|Sepp Hochreiter]] ('''1991'''). ''Untersuchungen zu dynamischen neuronalen Netzen''. Diploma thesis, [[Technical University of Munich|TU Munich]], advisor [[Jürgen Schmidhuber]], [http://people.idsia.ch/~juergen/SeppHochreiter1991ThesisAdvisorSchmidhuber.pdf pdf] (German) <ref>[http://people.idsia.ch/~juergen/fundamentaldeeplearningproblem.html Sepp Hochreiter's Fundamental Deep Learning Problem (1991)] by [[Jürgen Schmidhuber]], 2013</ref>
 
* [[Mathematician#SHochreiter|Sepp Hochreiter]] ('''1991'''). ''Untersuchungen zu dynamischen neuronalen Netzen''. Diploma thesis, [[Technical University of Munich|TU Munich]], advisor [[Jürgen Schmidhuber]], [http://people.idsia.ch/~juergen/SeppHochreiter1991ThesisAdvisorSchmidhuber.pdf pdf] (German) <ref>[http://people.idsia.ch/~juergen/fundamentaldeeplearningproblem.html Sepp Hochreiter's Fundamental Deep Learning Problem (1991)] by [[Jürgen Schmidhuber]], 2013</ref>
Line 212: Line 196:
 
* [[Justin A. Boyan]] ('''1992'''). ''Modular Neural Networks for Learning Context-Dependent Game Strategies''. Master's thesis, [https://en.wikipedia.org/wiki/University_of_Cambridge University of Cambridge], [http://www.cs.cmu.edu/~jab/cv/pubs/boyan.backgammon-thesis.pdf pdf]
 
* [[Justin A. Boyan]] ('''1992'''). ''Modular Neural Networks for Learning Context-Dependent Game Strategies''. Master's thesis, [https://en.wikipedia.org/wiki/University_of_Cambridge University of Cambridge], [http://www.cs.cmu.edu/~jab/cv/pubs/boyan.backgammon-thesis.pdf pdf]
 
* [https://en.wikipedia.org/wiki/Patricia_Churchland Patricia Churchland], [[Terrence J. Sejnowski]] ('''1992'''). ''[https://mitpress.mit.edu/books/computational-brain The Computational Brain]''. [https://en.wikipedia.org/wiki/MIT_Press MIT Press]  
 
* [https://en.wikipedia.org/wiki/Patricia_Churchland Patricia Churchland], [[Terrence J. Sejnowski]] ('''1992'''). ''[https://mitpress.mit.edu/books/computational-brain The Computational Brain]''. [https://en.wikipedia.org/wiki/MIT_Press MIT Press]  
 +
* [https://dblp.uni-trier.de/pers/hd/h/Hellstrom:Benjamin_J= Benjamin J. Hellstrom], [[Laveen Kanal|Laveen N. Kanal]] ('''1992'''). ''[https://ieeexplore.ieee.org/document/125871 Knapsack packing networks]''. [[IEEE#NN|IEEE Transactions on Neural Networks]], Vol. 3, No. 2
 +
* [https://dblp.uni-trier.de/pers/hd/h/Hellstrom:Benjamin_J= Benjamin J. Hellstrom], [[Laveen Kanal|Laveen N. Kanal]] ('''1992'''). ''Asymmetric mean-field neural networks for multiprocessor scheduling''. [https://en.wikipedia.org/wiki/Neural_Networks_(journal) Neural Networks], Vol. 5, No. 4
 
'''1993'''
 
'''1993'''
 
* [[Jacek Mańdziuk]], [http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/m/Macuk:Bohdan.html Bohdan Macukow] ('''1993'''). ''A Neural Network performing Boolean Logic Operations''. [http://www.springerlink.com/content/1060-992x/ Optical Memory and Neural Networks], Vol. 2, No. 1, [http://www.mini.pw.edu.pl/~mandziuk/PRACE/omnn93.pdf pdf]  
 
* [[Jacek Mańdziuk]], [http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/m/Macuk:Bohdan.html Bohdan Macukow] ('''1993'''). ''A Neural Network performing Boolean Logic Operations''. [http://www.springerlink.com/content/1060-992x/ Optical Memory and Neural Networks], Vol. 2, No. 1, [http://www.mini.pw.edu.pl/~mandziuk/PRACE/omnn93.pdf pdf]  
* [[Sebastian Thrun]], [[Tom Mitchell]] ('''1993'''). ''Integrating Inductive Neural Network Learning and Explanation-Based Learning''. Proceedings of the 13th IJCAI, pp. 930-936. Morgan Kaufmann, San Mateo, CA, [http://robots.stanford.edu/papers/thrun.EBNN_ijcai93.ps.gz zipped ps]
+
* [[Sebastian Thrun]], [[Tom Mitchell]] ('''1993'''). ''Integrating Inductive Neural Network Learning and Explanation-Based Learning''. Proceedings of the 13th IJCAI, Morgan Kaufmann, [http://robots.stanford.edu/papers/thrun.EBNN_ijcai93.ps.gz zipped ps]
 
* [[Byoung-Tak Zhang]], [[Mathematician#HMuehlenbein|Heinz Mühlenbein]] ('''1993'''). ''Evolving Optimal Neural Networks Using Genetic Algorithms with Occam's Razor''. [https://en.wikipedia.org/wiki/Complex_Systems_(journal) Complex Systems], Vol. 7, [http://www.complex-systems.com/pdf/07-3-2.pdf pdf]
 
* [[Byoung-Tak Zhang]], [[Mathematician#HMuehlenbein|Heinz Mühlenbein]] ('''1993'''). ''Evolving Optimal Neural Networks Using Genetic Algorithms with Occam's Razor''. [https://en.wikipedia.org/wiki/Complex_Systems_(journal) Complex Systems], Vol. 7, [http://www.complex-systems.com/pdf/07-3-2.pdf pdf]
 
* [[Martin Riedmiller]], [[Heinrich Braun]] ('''1993'''). ''A direct adaptive method for faster backpropagation learning: The RPROP algorithm''. [http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=1059 IEEE International Conference On Neural Networks], [http://paginas.fe.up.pt/~ee02162/dissertacao/RPROP%20paper.pdf pdf]
 
* [[Martin Riedmiller]], [[Heinrich Braun]] ('''1993'''). ''A direct adaptive method for faster backpropagation learning: The RPROP algorithm''. [http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=1059 IEEE International Conference On Neural Networks], [http://paginas.fe.up.pt/~ee02162/dissertacao/RPROP%20paper.pdf pdf]
 +
* [[Nicol N. Schraudolph]], [[Peter Dayan]], [[Terrence J. Sejnowski]] ('''1993'''). ''[https://papers.nips.cc/paper/820-temporal-difference-learning-of-position-evaluation-in-the-game-of-go Temporal Difference Learning of Position Evaluation in the Game of Go]''. [https://papers.nips.cc/book/advances-in-neural-information-processing-systems-6-1993 NIPS 1993] <ref>[http://satirist.org/learn-game/systems/go-net.html Nici Schraudolph’s go networks], review by [[Jay Scott]]</ref>
 
'''1994'''
 
'''1994'''
* [https://en.wikipedia.org/wiki/Paul_Werbos Paul Werbos] ('''1994'''). ''[http://eu.wiley.com/WileyCDA/WileyTitle/productCd-0471598976.html The Roots of Backpropagation. From Ordered Derivatives to Neural Networks and Political Forecasting]''. [https://en.wikipedia.org/wiki/John_Wiley_%26_Sons John Wiley & Sons]
+
* [[Mathematician#PWerbos|Paul Werbos]] ('''1994'''). ''The Roots of Backpropagation. From Ordered Derivatives to Neural Networks and Political Forecasting''. [https://en.wikipedia.org/wiki/John_Wiley_%26_Sons John Wiley & Sons]
 
* [[David E. Moriarty]], [[Risto Miikkulainen]] ('''1994'''). ''Evolving Neural Networks to focus Minimax Search''. [[AAAI|AAAI-94]], [http://www.cs.utexas.edu/~ai-lab/pubs/moriarty.focus.pdf pdf]
 
* [[David E. Moriarty]], [[Risto Miikkulainen]] ('''1994'''). ''Evolving Neural Networks to focus Minimax Search''. [[AAAI|AAAI-94]], [http://www.cs.utexas.edu/~ai-lab/pubs/moriarty.focus.pdf pdf]
 
* [[Eric Postma]] ('''1994'''). ''SCAN: A Neural Model of Covert Attention''. Ph.D. thesis, [[Maastricht University]], advisor [[Jaap van den Herik]]
 
* [[Eric Postma]] ('''1994'''). ''SCAN: A Neural Model of Covert Attention''. Ph.D. thesis, [[Maastricht University]], advisor [[Jaap van den Herik]]
 
* [[Sebastian Thrun]] ('''1994'''). ''Neural Network Learning in the Domain of Chess''. Machines That Learn, [http://snowbird.djvuzone.org/ Snowbird], Extended abstract
 
* [[Sebastian Thrun]] ('''1994'''). ''Neural Network Learning in the Domain of Chess''. Machines That Learn, [http://snowbird.djvuzone.org/ Snowbird], Extended abstract
 
* [[Christian Posthoff]], S. Schawelski, [[Michael Schlosser]] ('''1994'''). ''Neural Network Learning in a Chess Endgame Positions''. IEEE World Congress on Computational Intelligence
 
* [[Christian Posthoff]], S. Schawelski, [[Michael Schlosser]] ('''1994'''). ''Neural Network Learning in a Chess Endgame Positions''. IEEE World Congress on Computational Intelligence
* [[Nicol N. Schraudolph]], [[Peter Dayan]], [[Terrence J. Sejnowski]] ('''1994'''). ''[http://nic.schraudolph.org/bib2html/b2hd-SchDaySej94.html Temporal Difference Learning of Position Evaluation in the Game of Go]''. [http://papers.nips.cc/book/advances-in-neural-information-processing-systems-6-1993 Advances in Neural Information Processing Systems 6] <ref>[http://satirist.org/learn-game/systems/go-net.html Nici Schraudolph’s go networks], review by [[Jay Scott]]</ref>
 
 
* [[Alois Heinz]] ('''1994'''). ''[http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.55.3994 Efficient Neural Net α-β-Evaluators]''. [http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.55.3994&rep=rep1&type=pdf pdf] <ref>[https://www.stmintz.com/ccc/index.php?id=11893 Re: Evaluation by neural network ?] by [[Jay Scott]], [[CCC]], November 10, 1997</ref>
 
* [[Alois Heinz]] ('''1994'''). ''[http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.55.3994 Efficient Neural Net α-β-Evaluators]''. [http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.55.3994&rep=rep1&type=pdf pdf] <ref>[https://www.stmintz.com/ccc/index.php?id=11893 Re: Evaluation by neural network ?] by [[Jay Scott]], [[CCC]], November 10, 1997</ref>
 
* [[Alois Heinz]] ('''1994'''). ''Fast bounded smooth regression with lazy neural trees''. [http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=3013 ICNN 1994], DOI: [http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=374421 10.1109/ICNN.1994.374421]
 
* [[Alois Heinz]] ('''1994'''). ''Fast bounded smooth regression with lazy neural trees''. [http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=3013 ICNN 1994], DOI: [http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=374421 10.1109/ICNN.1994.374421]
 
* [[Martin Riedmiller]] ('''1994'''). ''Rprop - Description and Implementation Details''. Technical Report, [https://en.wikipedia.org/wiki/Karlsruhe_Institute_of_Technology University of Karlsruhe], [http://www.inf.fu-berlin.de/lehre/WS06/Musterererkennung/Paper/rprop.pdf pdf]
 
* [[Martin Riedmiller]] ('''1994'''). ''Rprop - Description and Implementation Details''. Technical Report, [https://en.wikipedia.org/wiki/Karlsruhe_Institute_of_Technology University of Karlsruhe], [http://www.inf.fu-berlin.de/lehre/WS06/Musterererkennung/Paper/rprop.pdf pdf]
 +
* [[Igor Kononenko]] ('''1994'''). ''On Bayesian Neural Networks''. [https://dblp.uni-trier.de/db/journals/informaticaSI/informaticaSI18.html Informatica (Slovenia), Vol. 18], No. 2
 
'''1995'''
 
'''1995'''
 
* [https://peterbraspenning.wordpress.com/ Peter J. Braspenning], [[Frank Thuijsman]], [https://scholar.google.com/citations?user=Ba9L7CAAAAAJ Ton Weijters] (eds) ('''1995'''). ''[http://link.springer.com/book/10.1007%2FBFb0027019 Artificial neural networks: an introduction to ANN theory and practice]''. [https://de.wikipedia.org/wiki/Lecture_Notes_in_Computer_Science LNCS] 931, [https://de.wikipedia.org/wiki/Springer_Science%2BBusiness_Media Springer]
 
* [https://peterbraspenning.wordpress.com/ Peter J. Braspenning], [[Frank Thuijsman]], [https://scholar.google.com/citations?user=Ba9L7CAAAAAJ Ton Weijters] (eds) ('''1995'''). ''[http://link.springer.com/book/10.1007%2FBFb0027019 Artificial neural networks: an introduction to ANN theory and practice]''. [https://de.wikipedia.org/wiki/Lecture_Notes_in_Computer_Science LNCS] 931, [https://de.wikipedia.org/wiki/Springer_Science%2BBusiness_Media Springer]
Line 251: Line 238:
 
* [[Pieter Spronck]] ('''1996'''). ''Elegance: Genetic Algorithms in Neural Reinforcement Control''. Master thesis, [[Delft University of Technology]], [http://ticc.uvt.nl/~pspronck/pubs/Elegance.pdf pdf]
 
* [[Pieter Spronck]] ('''1996'''). ''Elegance: Genetic Algorithms in Neural Reinforcement Control''. Master thesis, [[Delft University of Technology]], [http://ticc.uvt.nl/~pspronck/pubs/Elegance.pdf pdf]
 
* [[Raúl Rojas]] ('''1996'''). ''Neural Networks - A Systematic Introduction''. Springer, available as [http://www.inf.fu-berlin.de/inst/ag-ki/rojas_home/documents/1996/NeuralNetworks/neuron.pdf pdf ebook]
 
* [[Raúl Rojas]] ('''1996'''). ''Neural Networks - A Systematic Introduction''. Springer, available as [http://www.inf.fu-berlin.de/inst/ag-ki/rojas_home/documents/1996/NeuralNetworks/neuron.pdf pdf ebook]
 +
* [[Ida Sprinkhuizen-Kuyper]], [https://dblp.org/pers/hd/b/Boers:Egbert_J=_W= Egbert J. W. Boers] ('''1996'''). ''[https://ieeexplore.ieee.org/abstract/document/6796246 The Error Surface of the Simplest XOR Network Has Only Global Minima]''. [https://en.wikipedia.org/wiki/Neural_Computation_(journal) Neural Computation], Vol. 8, No. 6, [http://www.socsci.ru.nl/idak/publications/papers/NeuralComputation.pdf pdf]
 
'''1997'''
 
'''1997'''
 
* [[Mathematician#SHochreiter|Sepp Hochreiter]], [[Jürgen Schmidhuber]] ('''1997'''). ''Long short-term memory''. [https://en.wikipedia.org/wiki/Neural_Computation_%28journal%29 Neural Computation], Vol. 9, No. 8, [http://deeplearning.cs.cmu.edu/pdfs/Hochreiter97_lstm.pdf pdf] <ref>[https://en.wikipedia.org/wiki/Long_short_term_memory Long short term memory from Wikipedia]</ref>
 
* [[Mathematician#SHochreiter|Sepp Hochreiter]], [[Jürgen Schmidhuber]] ('''1997'''). ''Long short-term memory''. [https://en.wikipedia.org/wiki/Neural_Computation_%28journal%29 Neural Computation], Vol. 9, No. 8, [http://deeplearning.cs.cmu.edu/pdfs/Hochreiter97_lstm.pdf pdf] <ref>[https://en.wikipedia.org/wiki/Long_short_term_memory Long short term memory from Wikipedia]</ref>
Line 274: Line 262:
 
* [[Don Beal]], [[Martin C. Smith]] ('''1999'''). ''Learning Piece-Square Values using Temporal Differences.'' [[ICGA Journal#22_4|ICCA Journal, Vol. 22, No. 4]]
 
* [[Don Beal]], [[Martin C. Smith]] ('''1999'''). ''Learning Piece-Square Values using Temporal Differences.'' [[ICGA Journal#22_4|ICCA Journal, Vol. 22, No. 4]]
 
* [https://en.wikipedia.org/wiki/Simon_Haykin Simon S. Haykin] ('''1999'''). ''[http://dl.acm.org/citation.cfm?id=521706 Neural Networks: A Comprehensive Foundation]''. 2nd Edition, [https://en.wikipedia.org/wiki/Prentice_Hall Prentice-Hall]
 
* [https://en.wikipedia.org/wiki/Simon_Haykin Simon S. Haykin] ('''1999'''). ''[http://dl.acm.org/citation.cfm?id=521706 Neural Networks: A Comprehensive Foundation]''. 2nd Edition, [https://en.wikipedia.org/wiki/Prentice_Hall Prentice-Hall]
* [[Laurence F. Abbott]], [[Terrence J. Sejnowski]] (eds.) ('''1999'''). ''[https://mitpress.mit.edu/books/neural-codes-and-distributed-representations Neural Codes and Distributed Representations]''. [https://en.wikipedia.org/wiki/MIT_Press MIT Press]
+
* [https://en.wikipedia.org/wiki/Larry_Abbott Laurence F. Abbott], [[Terrence J. Sejnowski]] (eds.) ('''1999'''). ''[https://mitpress.mit.edu/books/neural-codes-and-distributed-representations Neural Codes and Distributed Representations]''. [https://en.wikipedia.org/wiki/MIT_Press MIT Press]
 
* [[Mathematician#GEHinton|Geoffrey E. Hinton]], [[Terrence J. Sejnowski]] (eds.) ('''1999'''). ''[https://mitpress.mit.edu/books/unsupervised-learning Unsupervised Learning: Foundations of Neural Computation]''.  [https://en.wikipedia.org/wiki/MIT_Press MIT Press]
 
* [[Mathematician#GEHinton|Geoffrey E. Hinton]], [[Terrence J. Sejnowski]] (eds.) ('''1999'''). ''[https://mitpress.mit.edu/books/unsupervised-learning Unsupervised Learning: Foundations of Neural Computation]''.  [https://en.wikipedia.org/wiki/MIT_Press MIT Press]
 
* [[Peter Dayan]] ('''1999'''). ''Recurrent Sampling Models for the Helmholtz Machine''. [https://en.wikipedia.org/wiki/Neural_Computation_(journal) Neural Computation], Vol. 11, No. 3, [http://www.gatsby.ucl.ac.uk/~dayan/papers/rechelm99.pdf pdf] <ref>[https://en.wikipedia.org/wiki/Helmholtz_machine Helmholtz machine from Wikipedia]</ref>
 
* [[Peter Dayan]] ('''1999'''). ''Recurrent Sampling Models for the Helmholtz Machine''. [https://en.wikipedia.org/wiki/Neural_Computation_(journal) Neural Computation], Vol. 11, No. 3, [http://www.gatsby.ucl.ac.uk/~dayan/papers/rechelm99.pdf pdf] <ref>[https://en.wikipedia.org/wiki/Helmholtz_machine Helmholtz machine from Wikipedia]</ref>
 +
* [[Ida Sprinkhuizen-Kuyper]], [https://dblp.org/pers/hd/b/Boers:Egbert_J=_W= Egbert J. W. Boers] ('''1999'''). ''[https://ieeexplore.ieee.org/document/774274 A local minimum for the 2-3-1 XOR network]''.  [[IEEE#NN|IEEE Transactions on Neural Networks]], Vol. 10, No. 4
 
==2000 ...==
 
==2000 ...==
 
* [[Levente Kocsis]], [[Jos Uiterwijk]], [[Jaap van den Herik]] ('''2000'''). ''[http://link.springer.com/chapter/10.1007%2F3-540-45579-5_11 Learning Time Allocation using Neural Networks]''. [[CG 2000]]
 
* [[Levente Kocsis]], [[Jos Uiterwijk]], [[Jaap van den Herik]] ('''2000'''). ''[http://link.springer.com/chapter/10.1007%2F3-540-45579-5_11 Learning Time Allocation using Neural Networks]''. [[CG 2000]]
Line 292: Line 281:
 
* [[Jonathan Schaeffer]], [[Markian Hlynka]], [[Vili Jussila]] ('''2001'''). ''Temporal Difference Learning Applied to a High-Performance Game-Playing Program''. [http://www.informatik.uni-trier.de/~ley/db/conf/ijcai/ijcai2001.html#SchaefferHJ01 IJCAI 2001]
 
* [[Jonathan Schaeffer]], [[Markian Hlynka]], [[Vili Jussila]] ('''2001'''). ''Temporal Difference Learning Applied to a High-Performance Game-Playing Program''. [http://www.informatik.uni-trier.de/~ley/db/conf/ijcai/ijcai2001.html#SchaefferHJ01 IJCAI 2001]
 
* [[Don Beal]], [[Martin C. Smith]] ('''2001'''). ''Temporal difference learning applied to game playing and the results of application to Shogi''. Theoretical Computer Science, Volume 252, Issues 1-2, pp. 105-119
 
* [[Don Beal]], [[Martin C. Smith]] ('''2001'''). ''Temporal difference learning applied to game playing and the results of application to Shogi''. Theoretical Computer Science, Volume 252, Issues 1-2, pp. 105-119
* [[Nicol N. Schraudolph]], [[Peter Dayan]], [[Terrence J. Sejnowski]]  ('''2001'''). ''[http://nic.schraudolph.org/bib2html/b2hd-SchDaySej01.html Learning to Evaluate Go Positions via Temporal Difference Methods]''. in  [[Norio Baba]], [[Lakhmi C. Jain]] (eds.) ('''2001'''). ''[http://jasss.soc.surrey.ac.uk/7/1/reviews/takama.html Computational Intelligence in Games, Studies in Fuzziness and Soft Computing]''. [http://www.springer.com/economics?SGWID=1-165-6-73481-0 Physica-Verlag]  
+
* [[Nicol N. Schraudolph]], [[Peter Dayan]], [[Terrence J. Sejnowski]]  ('''2001'''). ''[http://nic.schraudolph.org/bib2html/b2hd-SchDaySej01.html Learning to Evaluate Go Positions via Temporal Difference Methods]''. [http://jasss.soc.surrey.ac.uk/7/1/reviews/takama.html Computational Intelligence in Games, Studies in Fuzziness and Soft Computing], [http://www.springer.com/economics?SGWID=1-165-6-73481-0 Physica-Verlag]  
* [[Peter Dayan]], [[Laurence F. Abbott]] ('''2001, 2005'''). ''[http://www.gatsby.ucl.ac.uk/~dayan/book/index.html Theoretical Neuroscience: Computational and Mathematical Modeling of Neural Systems]''. [https://en.wikipedia.org/wiki/MIT_Press MIT Press]  
+
* [[Peter Dayan]], [https://en.wikipedia.org/wiki/Larry_Abbott Laurence F. Abbott] ('''2001, 2005'''). ''[http://www.gatsby.ucl.ac.uk/~dayan/book/index.html Theoretical Neuroscience: Computational and Mathematical Modeling of Neural Systems]''. [https://en.wikipedia.org/wiki/MIT_Press MIT Press]  
 
'''2002'''
 
'''2002'''
 
* [[Levente Kocsis]], [[Jos Uiterwijk]], [[Eric Postma]], [[Jaap van den Herik]] ('''2002'''). ''[http://link.springer.com/chapter/10.1007%2F978-3-540-40031-8_11 The Neural MoveMap Heuristic in Chess]''. [[CG 2002]]
 
* [[Levente Kocsis]], [[Jos Uiterwijk]], [[Eric Postma]], [[Jaap van den Herik]] ('''2002'''). ''[http://link.springer.com/chapter/10.1007%2F978-3-540-40031-8_11 The Neural MoveMap Heuristic in Chess]''. [[CG 2002]]
Line 302: Line 291:
 
* [[Moshe Sipper]] ('''2002''') ''[http://books.google.com/books/about/Machine_Nature.html?id=fbFQAAAAMAAJ&redir_esc=y Machine Nature: The Coming Age of Bio-Inspired Computing]''. [https://en.wikipedia.org/wiki/McGraw-Hill_Financial McGraw-Hill, New York]
 
* [[Moshe Sipper]] ('''2002''') ''[http://books.google.com/books/about/Machine_Nature.html?id=fbFQAAAAMAAJ&redir_esc=y Machine Nature: The Coming Age of Bio-Inspired Computing]''. [https://en.wikipedia.org/wiki/McGraw-Hill_Financial McGraw-Hill, New York]
 
* [[Paul E. Utgoff]], [[David J. Stracuzzi]] ('''2002'''). ''Many-Layered Learning''. [https://en.wikipedia.org/wiki/Neural_Computation_%28journal%29 Neural Computation], Vol. 14, No. 10, [http://people.cs.umass.edu/~utgoff/papers/neco-stl.pdf pdf]
 
* [[Paul E. Utgoff]], [[David J. Stracuzzi]] ('''2002'''). ''Many-Layered Learning''. [https://en.wikipedia.org/wiki/Neural_Computation_%28journal%29 Neural Computation], Vol. 14, No. 10, [http://people.cs.umass.edu/~utgoff/papers/neco-stl.pdf pdf]
* [[Michael I. Jordan]], [[Terrence J. Sejnowski]] (eds.) ('''2002'''). ''[https://mitpress.mit.edu/books/graphical-models Graphical Models: Foundations of Neural Computation]''. [https://en.wikipedia.org/wiki/MIT_Press MIT Press]
+
* [[Mathematician#MIJordan|Michael I. Jordan]], [[Terrence J. Sejnowski]] (eds.) ('''2002'''). ''[https://mitpress.mit.edu/books/graphical-models Graphical Models: Foundations of Neural Computation]''. [https://en.wikipedia.org/wiki/MIT_Press MIT Press]
 
'''2003'''
 
'''2003'''
 
* [[Levente Kocsis]] ('''2003'''). ''Learning Search Decisions''. Ph.D thesis, [[Maastricht University]], [https://project.dke.maastrichtuniversity.nl/games/files/phd/Kocsis_thesis.pdf pdf]
 
* [[Levente Kocsis]] ('''2003'''). ''Learning Search Decisions''. Ph.D thesis, [[Maastricht University]], [https://project.dke.maastrichtuniversity.nl/games/files/phd/Kocsis_thesis.pdf pdf]
Line 366: Line 355:
 
* [[Matthew Lai]] ('''2015'''). ''Giraffe: Using Deep Reinforcement Learning to Play Chess''. M.Sc. thesis, [https://en.wikipedia.org/wiki/Imperial_College_London Imperial College London],  [http://arxiv.org/abs/1509.01549v1 arXiv:1509.01549v1] » [[Giraffe]]
 
* [[Matthew Lai]] ('''2015'''). ''Giraffe: Using Deep Reinforcement Learning to Play Chess''. M.Sc. thesis, [https://en.wikipedia.org/wiki/Imperial_College_London Imperial College London],  [http://arxiv.org/abs/1509.01549v1 arXiv:1509.01549v1] » [[Giraffe]]
 
* [[Nikolai Yakovenko]], [[Liangliang Cao]], [[Colin Raffel]], [[James Fan]] ('''2015'''). ''Poker-CNN: A Pattern Learning Strategy for Making Draws and Bets in Poker Games''. [https://arxiv.org/abs/1509.06731 arXiv:1509.06731]
 
* [[Nikolai Yakovenko]], [[Liangliang Cao]], [[Colin Raffel]], [[James Fan]] ('''2015'''). ''Poker-CNN: A Pattern Learning Strategy for Making Draws and Bets in Poker Games''. [https://arxiv.org/abs/1509.06731 arXiv:1509.06731]
 +
* [[Ilya Loshchilov]], [[Frank Hutter]] ('''2015'''). ''Online Batch Selection for Faster Training of Neural Networks''. [https://arxiv.org/abs/1511.06343 arXiv:1511.06343]
 
* [[Yuandong Tian]], [[Yan Zhu]] ('''2015'''). ''Better Computer Go Player with Neural Network and Long-term Prediction''. [http://arxiv.org/abs/1511.06410 arXiv:1511.06410] <ref>[http://www.technologyreview.com/view/544181/how-facebooks-ai-researchers-built-a-game-changing-go-engine/?utm_campaign=socialsync&utm_medium=social-post&utm_source=facebook How Facebook’s AI Researchers Built a Game-Changing Go Engine | MIT Technology Review], December 04, 2015</ref> <ref>[http://www.talkchess.com/forum/viewtopic.php?t=58514 Combining Neural Networks and Search techniques (GO)] by Michael Babigian, [[CCC]], December 08, 2015</ref> » [[Go]]
 
* [[Yuandong Tian]], [[Yan Zhu]] ('''2015'''). ''Better Computer Go Player with Neural Network and Long-term Prediction''. [http://arxiv.org/abs/1511.06410 arXiv:1511.06410] <ref>[http://www.technologyreview.com/view/544181/how-facebooks-ai-researchers-built-a-game-changing-go-engine/?utm_campaign=socialsync&utm_medium=social-post&utm_source=facebook How Facebook’s AI Researchers Built a Game-Changing Go Engine | MIT Technology Review], December 04, 2015</ref> <ref>[http://www.talkchess.com/forum/viewtopic.php?t=58514 Combining Neural Networks and Search techniques (GO)] by Michael Babigian, [[CCC]], December 08, 2015</ref> » [[Go]]
 
* [[Peter H. Jin]], [[Kurt Keutzer]] ('''2015'''). ''Convolutional Monte Carlo Rollouts in Go''. [http://arxiv.org/abs/1512.03375 arXiv:1512.03375]
 
* [[Peter H. Jin]], [[Kurt Keutzer]] ('''2015'''). ''Convolutional Monte Carlo Rollouts in Go''. [http://arxiv.org/abs/1512.03375 arXiv:1512.03375]
Line 384: Line 374:
 
* [https://scholar.google.ca/citations?user=mZfgLA4AAAAJ&hl=en Vincent Dumoulin], [https://scholar.google.it/citations?user=kaAnZw0AAAAJ&hl=en Francesco Visin] ('''2016'''). ''A guide to convolution arithmetic for deep learning''. [https://arxiv.org/abs/1603.07285 arXiv:1603.07285]
 
* [https://scholar.google.ca/citations?user=mZfgLA4AAAAJ&hl=en Vincent Dumoulin], [https://scholar.google.it/citations?user=kaAnZw0AAAAJ&hl=en Francesco Visin] ('''2016'''). ''A guide to convolution arithmetic for deep learning''. [https://arxiv.org/abs/1603.07285 arXiv:1603.07285]
 
* [https://en.wikipedia.org/wiki/Patricia_Churchland Patricia Churchland], [[Terrence J. Sejnowski]] ('''2016'''). ''[https://mitpress.mit.edu/books/computational-brain-0 The Computational Brain, 25th Anniversary Edition]''. [https://en.wikipedia.org/wiki/MIT_Press MIT Press]  
 
* [https://en.wikipedia.org/wiki/Patricia_Churchland Patricia Churchland], [[Terrence J. Sejnowski]] ('''2016'''). ''[https://mitpress.mit.edu/books/computational-brain-0 The Computational Brain, 25th Anniversary Edition]''. [https://en.wikipedia.org/wiki/MIT_Press MIT Press]  
 +
* [[Ilya Loshchilov]], [[Frank Hutter]] ('''2016'''). ''CMA-ES for Hyperparameter Optimization of Deep Neural Networks''. [https://arxiv.org/abs/1604.07269 arXiv:1604.07269] <ref>[https://en.wikipedia.org/wiki/CMA-ES CMA-ES from Wikipedia]</ref>
 
* [[Audrūnas Gruslys]], [[Rémi Munos]], [[Ivo Danihelka]], [[Marc Lanctot]], [[Alex Graves]] ('''2016'''). ''Memory-Efficient Backpropagation Through Time''. [https://arxiv.org/abs/1606.03401v1 arXiv:1606.03401]
 
* [[Audrūnas Gruslys]], [[Rémi Munos]], [[Ivo Danihelka]], [[Marc Lanctot]], [[Alex Graves]] ('''2016'''). ''Memory-Efficient Backpropagation Through Time''. [https://arxiv.org/abs/1606.03401v1 arXiv:1606.03401]
 
* [[Andrei A. Rusu]], [[Neil C. Rabinowitz]], [[Guillaume Desjardins]], [[Hubert Soyer]], [[James Kirkpatrick]], [[Koray Kavukcuoglu]], [[Razvan Pascanu]], [[Raia Hadsell]] ('''2016'''). ''Progressive Neural Networks''. [https://arxiv.org/abs/1606.04671 arXiv:1606.04671]
 
* [[Andrei A. Rusu]], [[Neil C. Rabinowitz]], [[Guillaume Desjardins]], [[Hubert Soyer]], [[James Kirkpatrick]], [[Koray Kavukcuoglu]], [[Razvan Pascanu]], [[Raia Hadsell]] ('''2016'''). ''Progressive Neural Networks''. [https://arxiv.org/abs/1606.04671 arXiv:1606.04671]
 
* [[George Rajna]] ('''2016'''). ''Deep Neural Networks''. [http://vixra.org/abs/1609.0126 viXra:1609.0126]
 
* [[George Rajna]] ('''2016'''). ''Deep Neural Networks''. [http://vixra.org/abs/1609.0126 viXra:1609.0126]
* [[James Kirkpatrick]], [[Razvan Pascanu]], [[Neil C. Rabinowitz]], [[Joel Veness]], [[Guillaume Desjardins]], [[Andrei A. Rusu]], [[Kieran Milan]], [[John Quan]], [[Tiago Ramalho]],  [[Agnieszka Grabska-Barwinska]], [[Demis Hassabis]], [[Claudia Clopath]], [[Dharshan Kumaran]], [[Raia Hadsell]] ('''2016'''). ''Overcoming catastrophic forgetting in neural networks''. [https://arxiv.org/abs/1612.00796 arXiv:1612.00796]
+
* [[James Kirkpatrick]], [[Razvan Pascanu]], [[Neil C. Rabinowitz]], [[Joel Veness]], [[Guillaume Desjardins]], [[Andrei A. Rusu]], [[Kieran Milan]], [[John Quan]], [[Tiago Ramalho]],  [[Agnieszka Grabska-Barwinska]], [[Demis Hassabis]], [[Claudia Clopath]], [[Dharshan Kumaran]], [[Raia Hadsell]] ('''2016'''). ''Overcoming catastrophic forgetting in neural networks''. [https://arxiv.org/abs/1612.00796 arXiv:1612.00796] <ref>[http://www.talkchess.com/forum3/viewtopic.php?f=7&t=70704 catastrophic forgetting] by [[Daniel Shawul]], [[CCC]], May 09, 2019</ref>
 +
* [https://dblp.uni-trier.de/pers/hd/n/Niu:Zhenxing Zhenxing Niu], [https://dblp.uni-trier.de/pers/hd/z/Zhou:Mo Mo Zhou], [https://dblp.uni-trier.de/pers/hd/w/Wang_0003:Le Le Wang], [[Xinbo Gao]], [https://dblp.uni-trier.de/pers/hd/h/Hua_0001:Gang Gang Hua] ('''2016'''). ''Ordinal Regression with Multiple Output CNN for Age Estimation''. [https://dblp.uni-trier.de/db/conf/cvpr/cvpr2016.html CVPR 2016], [https://www.cv-foundation.org/openaccess/content_cvpr_2016/app/S21-20.pdf pdf]
 
'''2017'''
 
'''2017'''
 
* [[Yutian Chen]], [[Matthew W. Hoffman]], [[Sergio Gomez Colmenarejo]], [[Misha Denil]], [[Timothy Lillicrap]], [[Matthew Botvinick]], [[Nando de Freitas]] ('''2017'''). ''Learning to Learn without Gradient Descent by Gradient Descent''. [https://arxiv.org/abs/1611.03824v6 arXiv:1611.03824v6], [http://dblp.uni-trier.de/db/conf/icml/icml2017.html ICML 2017]
 
* [[Yutian Chen]], [[Matthew W. Hoffman]], [[Sergio Gomez Colmenarejo]], [[Misha Denil]], [[Timothy Lillicrap]], [[Matthew Botvinick]], [[Nando de Freitas]] ('''2017'''). ''Learning to Learn without Gradient Descent by Gradient Descent''. [https://arxiv.org/abs/1611.03824v6 arXiv:1611.03824v6], [http://dblp.uni-trier.de/db/conf/icml/icml2017.html ICML 2017]
Line 395: Line 387:
 
* [[Raúl Rojas]] ('''2017'''). ''Deepest Neural Networks''. [https://arxiv.org/abs/1707.02617 arXiv:1707.02617]
 
* [[Raúl Rojas]] ('''2017'''). ''Deepest Neural Networks''. [https://arxiv.org/abs/1707.02617 arXiv:1707.02617]
 
* [[Matej Moravčík]], [[Martin Schmid]], [[Neil Burch]], [[Viliam Lisý]], [[Dustin Morrill]], [[Nolan Bard]], [[Trevor Davis]], [[Kevin Waugh]], [[Michael Johanson]], [[Michael Bowling]] ('''2017'''). ''[http://science.sciencemag.org/content/356/6337/508 DeepStack: Expert-level artificial intelligence in heads-up no-limit poker]''. [https://en.wikipedia.org/wiki/Science_(journal) Science], Vol. 356, No. 6337
 
* [[Matej Moravčík]], [[Martin Schmid]], [[Neil Burch]], [[Viliam Lisý]], [[Dustin Morrill]], [[Nolan Bard]], [[Trevor Davis]], [[Kevin Waugh]], [[Michael Johanson]], [[Michael Bowling]] ('''2017'''). ''[http://science.sciencemag.org/content/356/6337/508 DeepStack: Expert-level artificial intelligence in heads-up no-limit poker]''. [https://en.wikipedia.org/wiki/Science_(journal) Science], Vol. 356, No. 6337
 +
* [[Xinqi Zhu]], [[Michael Bain]] ('''2017'''). ''B-CNN: Branch Convolutional Neural Network for Hierarchical Classification''. [https://arxiv.org/abs/1709.09890 arXiv:1709.09890], [https://github.com/zhuxinqimac/B-CNN GitHub - zhuxinqimac/B-CNN: Sample code of B-CNN paper]
 
* [[David Silver]], [[Julian Schrittwieser]], [[Karen Simonyan]], [[Ioannis Antonoglou]], [[Shih-Chieh Huang|Aja Huang]], [[Arthur Guez]], [[Thomas Hubert]], [[Lucas Baker]], [[Matthew Lai]], [[Adrian Bolton]], [[Yutian Chen]], [[Timothy Lillicrap]], [[Fan Hui]], [[Laurent Sifre]], [[George van den Driessche]], [[Thore Graepel]], [[Demis Hassabis]] ('''2017'''). ''[https://www.nature.com/nature/journal/v550/n7676/full/nature24270.html Mastering the game of Go without human knowledge]''. [https://en.wikipedia.org/wiki/Nature_%28journal%29 Nature], Vol. 550 <ref>[https://deepmind.com/blog/alphago-zero-learning-scratch/ AlphaGo Zero: Learning from scratch] by [[Demis Hassabis]] and [[David Silver]], [[DeepMind]], October 18, 2017</ref>
 
* [[David Silver]], [[Julian Schrittwieser]], [[Karen Simonyan]], [[Ioannis Antonoglou]], [[Shih-Chieh Huang|Aja Huang]], [[Arthur Guez]], [[Thomas Hubert]], [[Lucas Baker]], [[Matthew Lai]], [[Adrian Bolton]], [[Yutian Chen]], [[Timothy Lillicrap]], [[Fan Hui]], [[Laurent Sifre]], [[George van den Driessche]], [[Thore Graepel]], [[Demis Hassabis]] ('''2017'''). ''[https://www.nature.com/nature/journal/v550/n7676/full/nature24270.html Mastering the game of Go without human knowledge]''. [https://en.wikipedia.org/wiki/Nature_%28journal%29 Nature], Vol. 550 <ref>[https://deepmind.com/blog/alphago-zero-learning-scratch/ AlphaGo Zero: Learning from scratch] by [[Demis Hassabis]] and [[David Silver]], [[DeepMind]], October 18, 2017</ref>
 
* [[David Silver]], [[Thomas Hubert]], [[Julian Schrittwieser]], [[Ioannis Antonoglou]], [[Matthew Lai]], [[Arthur Guez]], [[Marc Lanctot]], [[Laurent Sifre]], [[Dharshan Kumaran]], [[Thore Graepel]], [[Timothy Lillicrap]], [[Karen Simonyan]], [[Demis Hassabis]] ('''2017'''). ''Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm''. [https://arxiv.org/abs/1712.01815 arXiv:1712.01815]  » [[Neural Networks#AlphaZero|AlphaZero]] <ref>[http://www.talkchess.com/forum/viewtopic.php?t=65909 Google's AlphaGo team has been working on chess] by [[Peter Kappler]], [[CCC]], December 06, 2017</ref>
 
* [[David Silver]], [[Thomas Hubert]], [[Julian Schrittwieser]], [[Ioannis Antonoglou]], [[Matthew Lai]], [[Arthur Guez]], [[Marc Lanctot]], [[Laurent Sifre]], [[Dharshan Kumaran]], [[Thore Graepel]], [[Timothy Lillicrap]], [[Karen Simonyan]], [[Demis Hassabis]] ('''2017'''). ''Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm''. [https://arxiv.org/abs/1712.01815 arXiv:1712.01815]  » [[Neural Networks#AlphaZero|AlphaZero]] <ref>[http://www.talkchess.com/forum/viewtopic.php?t=65909 Google's AlphaGo team has been working on chess] by [[Peter Kappler]], [[CCC]], December 06, 2017</ref>
Line 400: Line 393:
 
* [[Kei Takada]], [[Hiroyuki Iizuka]], [[Masahito Yamamoto]] ('''2017'''). ''Reinforcement Learning for Creating Evaluation Function Using Convolutional Neural Network in Hex''. TAAI 2017 » [[Hex]]
 
* [[Kei Takada]], [[Hiroyuki Iizuka]], [[Masahito Yamamoto]] ('''2017'''). ''Reinforcement Learning for Creating Evaluation Function Using Convolutional Neural Network in Hex''. TAAI 2017 » [[Hex]]
 
* [[Chao Gao]], [[Martin Müller]], [[Ryan Hayward]] ('''2017'''). ''Focused Depth-first Proof Number Search using Convolutional Neural Networks for the Game of Hex''. [[Conferences#IJCAI2017|IJCAI 2017]]
 
* [[Chao Gao]], [[Martin Müller]], [[Ryan Hayward]] ('''2017'''). ''Focused Depth-first Proof Number Search using Convolutional Neural Networks for the Game of Hex''. [[Conferences#IJCAI2017|IJCAI 2017]]
 +
* [[Thomas Elsken]], [[Jan Hendrik Metzen]], [[Frank Hutter]] ('''2017'''). ''Simple And Efficient Architecture Search for Convolutional Neural Networks''. [https://arxiv.org/abs/1711.04528 arXiv:1711.04528]
 +
* [[Joel Veness]], [[Tor Lattimore]], [https://github.com/avishkar58 Avishkar Bhoopchand], [https://scholar.google.co.uk/citations?user=mB4yebIAAAAJ&hl=en Agnieszka Grabska-Barwinska], [https://dblp.org/pers/hd/m/Mattern:Christopher Christopher Mattern], [https://dblp.org/pers/hd/t/Toth:Peter Peter Toth] ('''2017'''). ''Online Learning with Gated Linear Networks''. [https://arxiv.org/abs/1712.01897 arXiv:1712.01897]
 
* [https://dblp.uni-trier.de/pers/hd/c/Chen:Qiming Qiming Chen], [[Ren Wu]] ('''2017'''). ''CNN Is All You Need''. [https://arxiv.org/abs/1712.09662 arXiv:1712.09662]
 
* [https://dblp.uni-trier.de/pers/hd/c/Chen:Qiming Qiming Chen], [[Ren Wu]] ('''2017'''). ''CNN Is All You Need''. [https://arxiv.org/abs/1712.09662 arXiv:1712.09662]
 +
* [https://dblp.org/pers/hd/s/Serb:Alexander Alexantrou Serb], [[Edoardo Manino]], [https://dblp.org/pers/hd/m/Messaris:Ioannis Ioannis Messaris], [https://dblp.org/pers/hd/t/Tran=Thanh:Long Long Tran-Thanh], [https://www.orc.soton.ac.uk/people/tp1f12 Themis Prodromakis] ('''2017'''). ''[https://eprints.soton.ac.uk/425616/ Hardware-level Bayesian inference]''. [https://nips.cc/Conferences/2017 NIPS 2017] » [[Analog Evaluation]]
 
'''2018'''
 
'''2018'''
 
* [[Kei Takada]], [[Hiroyuki Iizuka]], [[Masahito Yamamoto]] ('''2018'''). ''[https://link.springer.com/chapter/10.1007%2F978-3-319-75931-9_2 Computer Hex Algorithm Using a Move Evaluation Method Based on a Convolutional Neural Network]''. [https://link.springer.com/bookseries/7899 Communications in Computer and Information Science] » [[Hex]]
 
* [[Kei Takada]], [[Hiroyuki Iizuka]], [[Masahito Yamamoto]] ('''2018'''). ''[https://link.springer.com/chapter/10.1007%2F978-3-319-75931-9_2 Computer Hex Algorithm Using a Move Evaluation Method Based on a Convolutional Neural Network]''. [https://link.springer.com/bookseries/7899 Communications in Computer and Information Science] » [[Hex]]
 
* [[Matthia Sabatelli]], [[Francesco Bidoia]], [[Valeriu Codreanu]], [[Marco Wiering]] ('''2018'''). ''Learning to Evaluate Chess Positions with Deep Neural Networks and Limited Lookahead''. ICPRAM 2018, [http://www.ai.rug.nl/~mwiering/GROUP/ARTICLES/ICPRAM_CHESS_DNN_2018.pdf pdf]
 
* [[Matthia Sabatelli]], [[Francesco Bidoia]], [[Valeriu Codreanu]], [[Marco Wiering]] ('''2018'''). ''Learning to Evaluate Chess Positions with Deep Neural Networks and Limited Lookahead''. ICPRAM 2018, [http://www.ai.rug.nl/~mwiering/GROUP/ARTICLES/ICPRAM_CHESS_DNN_2018.pdf pdf]
 +
* [[Ashwin Srinivasan]], [[Lovekesh Vig]], [[Michael Bain]] ('''2018'''). ''Logical Explanations for Deep Relational Machines Using Relevance Information''. [https://arxiv.org/abs/1807.00595 arXiv:1807.00595]
 +
* [[Thomas Elsken]], [[Jan Hendrik Metzen]], [[Frank Hutter]] ('''2018'''). ''Neural Architecture Search: A Survey''. [https://arxiv.org/abs/1808.05377 arXiv:1808.05377]
 
* [[David Silver]], [[Thomas Hubert]], [[Julian Schrittwieser]], [[Ioannis Antonoglou]], [[Matthew Lai]], [[Arthur Guez]], [[Marc Lanctot]], [[Laurent Sifre]], [[Dharshan Kumaran]], [[Thore Graepel]], [[Timothy Lillicrap]], [[Karen Simonyan]], [[Demis Hassabis]] ('''2018'''). ''[http://science.sciencemag.org/content/362/6419/1140 A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play]''. [https://en.wikipedia.org/wiki/Science_(journal) Science], Vol. 362, No. 6419 <ref>[https://deepmind.com/blog/alphazero-shedding-new-light-grand-games-chess-shogi-and-go/ AlphaZero: Shedding new light on the grand games of chess, shogi and Go] by [[David Silver]], [[Thomas Hubert]], [[Julian Schrittwieser]] and [[Demis Hassabis]], [[DeepMind]], December 03, 2018</ref>
 
* [[David Silver]], [[Thomas Hubert]], [[Julian Schrittwieser]], [[Ioannis Antonoglou]], [[Matthew Lai]], [[Arthur Guez]], [[Marc Lanctot]], [[Laurent Sifre]], [[Dharshan Kumaran]], [[Thore Graepel]], [[Timothy Lillicrap]], [[Karen Simonyan]], [[Demis Hassabis]] ('''2018'''). ''[http://science.sciencemag.org/content/362/6419/1140 A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play]''. [https://en.wikipedia.org/wiki/Science_(journal) Science], Vol. 362, No. 6419 <ref>[https://deepmind.com/blog/alphazero-shedding-new-light-grand-games-chess-shogi-and-go/ AlphaZero: Shedding new light on the grand games of chess, shogi and Go] by [[David Silver]], [[Thomas Hubert]], [[Julian Schrittwieser]] and [[Demis Hassabis]], [[DeepMind]], December 03, 2018</ref>
 +
* [[Chao Gao]], [[Siqi Yan]], [[Ryan Hayward]], [[Martin Müller]] ('''2018'''). ''A transferable neural network for Hex''. [[ICGA Journal#40_3|ICGA Journal, Vol. 40, No. 3]]
 +
'''2019'''
 +
* [[Marius Lindauer]], [[Frank Hutter]] ('''2019'''). ''Best Practices for Scientific Research on Neural Architecture Search''. [https://arxiv.org/abs/1909.02453 arXiv:1909.02453]
  
 
=Blog & Forum Posts=
 
=Blog & Forum Posts=
Line 458: Line 459:
 
* [http://www.talkchess.com/forum/viewtopic.php?t=64096 Is AlphaGo approach unsuitable to chess?] by Mel Cooper, [[CCC]], May 27, 2017 » [[AlphaGo]], [[Deep Learning]], [[Giraffe]]
 
* [http://www.talkchess.com/forum/viewtopic.php?t=64096 Is AlphaGo approach unsuitable to chess?] by Mel Cooper, [[CCC]], May 27, 2017 » [[AlphaGo]], [[Deep Learning]], [[Giraffe]]
 
: [http://www.talkchess.com/forum/viewtopic.php?t=64096&start=12 Re: Is AlphaGo approach unsuitable to chess?] by [[Peter Österlund]], [[CCC]], May 31, 2017 » [[Texel]]
 
: [http://www.talkchess.com/forum/viewtopic.php?t=64096&start=12 Re: Is AlphaGo approach unsuitable to chess?] by [[Peter Österlund]], [[CCC]], May 31, 2017 » [[Texel]]
* [https://groups.google.com/d/msg/computer-go-archive/WImAk15gRN4/bhA7kSAnBgAJ Neural nets for Go - chain pooling?] by [[David Wu]], [https://groups.google.com/forum/#!forum/computer-go-archive Computer Go Archive], August 18, 2017
+
* [https://groups.google.com/d/msg/computer-go-archive/WImAk15gRN4/bhA7kSAnBgAJ Neural nets for Go - chain pooling?] by [[David J. Wu|David Wu]], [https://groups.google.com/forum/#!forum/computer-go-archive Computer Go Archive], August 18, 2017
 
* [https://deepmind.com/blog/alphago-zero-learning-scratch/ AlphaGo Zero: Learning from scratch] by [[Demis Hassabis]] and [[David Silver]], [[DeepMind]], October 18, 2017
 
* [https://deepmind.com/blog/alphago-zero-learning-scratch/ AlphaGo Zero: Learning from scratch] by [[Demis Hassabis]] and [[David Silver]], [[DeepMind]], October 18, 2017
 
* [http://www.talkchess.com/forum/viewtopic.php?t=65481 We are doomed - AlphaGo Zero, learning only from basic rules] by [[Vincent Lejeune]], [[CCC]], October 18, 2017
 
* [http://www.talkchess.com/forum/viewtopic.php?t=65481 We are doomed - AlphaGo Zero, learning only from basic rules] by [[Vincent Lejeune]], [[CCC]], October 18, 2017
Line 475: Line 476:
 
* [http://www.talkchess.com/forum/viewtopic.php?t=66681 3 million games for training neural networks] by [[Álvaro Begué]], [[CCC]], February 24, 2018 » [[Automated Tuning]]
 
* [http://www.talkchess.com/forum/viewtopic.php?t=66681 3 million games for training neural networks] by [[Álvaro Begué]], [[CCC]], February 24, 2018 » [[Automated Tuning]]
 
* [http://www.talkchess.com/forum/viewtopic.php?t=66791 Looking inside NNs] by [[J. Wesley Cleveland]], [[CCC]], March 09, 2018
 
* [http://www.talkchess.com/forum/viewtopic.php?t=66791 Looking inside NNs] by [[J. Wesley Cleveland]], [[CCC]], March 09, 2018
 +
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=67347 GPU ANN, how to deal with host-device latencies?] by [[Srdja Matovic]], [[CCC]], May 06, 2018 » [[GPU]]
 
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=67524 Poor man's neurones] by [[Pawel Koziol]], [[CCC]], May 21, 2018 » [[Evaluation]]
 
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=67524 Poor man's neurones] by [[Pawel Koziol]], [[CCC]], May 21, 2018 » [[Evaluation]]
 
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=67600 Egbb dll neural network support] by [[Daniel Shawul]], [[CCC]], May 29, 2018 » [[Scorpio Bitbases]]
 
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=67600 Egbb dll neural network support] by [[Daniel Shawul]], [[CCC]], May 29, 2018 » [[Scorpio Bitbases]]
Line 480: Line 482:
 
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=69069 Are draws hard to predict?] by [[Daniel Shawul]], [[CCC]], November 27, 2018 » [[Draw]]
 
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=69069 Are draws hard to predict?] by [[Daniel Shawul]], [[CCC]], November 27, 2018 » [[Draw]]
 
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=69393 neural network architecture] by jackd, [[CCC]], December 26, 2018
 
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=69393 neural network architecture] by jackd, [[CCC]], December 26, 2018
 +
'''2019'''
 +
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=69795 So, how many of you are working on neural networks for chess?] by [[Srdja Matovic]], [[CCC]], February 01, 2019
 +
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=69942 categorical cross entropy for value] by [[Chris Whittington]], [[CCC]], February 18, 2019
 +
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=70504 Google's bfloat for neural networks] by [[Srdja Matovic]], [[CCC]], April 16, 2019 » [[Float]]
 +
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=70704 catastrophic forgetting] by [[Daniel Shawul]], [[CCC]], May 09, 2019 » [[Nebiyu]]
 +
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=71269 Wouldn’t it be nice if there was a ChessNet50] by [[Chris Whittington]], [[CCC]], July 13, 2019
 +
* [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=71301 A question to MCTS + NN experts] by [[Maksim Korzh]], [[CCC]], July 17, 2019 » [[Monte-Carlo Tree Search]]
 +
: [http://www.talkchess.com/forum3/viewtopic.php?f=7&t=71301&start=3 Re: A question to MCTS + NN experts] by [[Daniel Shawul]], [[CCC]], July 17, 2019
  
 
=External Links=
 
=External Links=
Line 513: Line 523:
 
* [https://en.wikipedia.org/wiki/Neocognitron Neocognitron from Wikipedia]
 
* [https://en.wikipedia.org/wiki/Neocognitron Neocognitron from Wikipedia]
 
* [http://www.scholarpedia.org/article/Neocognitron Neocognitron - Scholarpedia] by [http://www.scholarpedia.org/article/User:Kunihiko_Fukushima Kunihiko Fukushima]
 
* [http://www.scholarpedia.org/article/Neocognitron Neocognitron - Scholarpedia] by [http://www.scholarpedia.org/article/User:Kunihiko_Fukushima Kunihiko Fukushima]
 +
* [https://en.wikipedia.org/wiki/Neural_architecture_search Neural architecture search from Wikipedia]
 
* [https://en.wikipedia.org/wiki/Neuromorphic_engineering Neuromorphic engineering from Wikipedia]
 
* [https://en.wikipedia.org/wiki/Neuromorphic_engineering Neuromorphic engineering from Wikipedia]
 
: [https://en.wikipedia.org/wiki/Neurogrid Neurogrid from Wikipedia]
 
: [https://en.wikipedia.org/wiki/Neurogrid Neurogrid from Wikipedia]

Revision as of 21:13, 6 January 2020

Home * Learning * Neural Networks

Artificial Neural Network [1]

Neural Networks,
a series of connected neurons which communicate due to neurotransmission. The interface through which neurons interact with their neighbors consists of axon terminals connected via synapses to dendrites on other neurons. If the sum of the input signals into one neuron surpasses a certain threshold, the neuron sends an action potential at the axon hillock and transmits this electrical signal along the axon.

In 1949, Donald O. Hebb introduced his theory in The Organization of Behavior, stating that learning is about to adapt weight vectors (persistent synaptic plasticity) of the neuron pre-synaptic inputs, whose dot-product activates or controls the post-synaptic output, which is the base of Neural network learning [2].

AN

Already in the early 40s, Warren S. McCulloch and Walter Pitts introduced the artificial neuron as a logical element with multiple analogue inputs and a single digital output with a boolean result. The output fired "true", if the sum of the inputs exceed a threshold. In their 1943 paper A Logical Calculus of the Ideas Immanent in Nervous Activity [3], they attempted to demonstrate that a Turing machine program could be implemented in a finite network of such neurons of combinatorial logic functions of AND, OR and NOT.

ANNs

Artificial Neural Networks (ANNs) are a family of statistical learning devices or algorithms used in regression, and binary or multiclass classification, implemented in hardware or software inspired by their biological counterparts. The artificial neurons of one or more layers receive one or more inputs (representing dendrites), and after being weighted, sum them to produce an output (representing a neuron's axon). The sum is passed through a nonlinear function known as an activation function or transfer function. The transfer functions usually have a sigmoid shape, but they may also take the form of other non-linear functions, piecewise linear functions, or step functions [4]. The weights of the inputs of each layer are tuned to minimize a cost or loss function, which is a task in mathematical optimization and machine learning.

Perceptron

Perceptron [5]

The perceptron is an algorithm for supervised learning of binary classifiers. It was the first artificial neural network, introduced in 1957 by Frank Rosenblatt [6], implemented in custom hardware. In its basic form it consists of a single neuron with multiple inputs and associated weights.

Supervised learning is applied using a set D of labeled training data with pairs of feature vectors (x) and given results as desired output (d), usually started with cleared or randomly initialized weight vector w. The output is calculated by all inputs of a sample, multiplied by its corresponding weights, passing the sum to the activation function f. The difference of desired and actual value is then immediately used modify the weights for all features using a learning rate 0.0 < α <= 1.0:

   for (j=0, Σ = 0.0; j < nSamples; ++j) {
    for (i=0, X = bias; i < nFeatures; ++i) 
      X += w[i]*x[j][i];
    y = f ( X );
    Σ += abs(Δ = d[j] - y);
    for (i=0; i < nFeatures; ++i) 
      w[i] += α*Δ*x[j][i];
   }

AI Winter

Three layer, XOR capable Perceptron [7]

Although the perceptron initially seemed promising, it was proved that perceptrons could not be trained to recognise many classes of patterns. This led to neural network research stagnating for many years, the AI-winter, before it was recognised that a feedforward neural network with two or more layers had far greater processing power than with one layer. Single layer perceptrons are only capable of learning linearly separable patterns. In their 1969 book Perceptrons, Marvin Minsky and Seymour Papert wrote that it was impossible for these classes of network to learn the XOR function. It is often believed that they also conjectured (incorrectly) that a similar result would hold for a multilayer perceptron [8]. However, this is not true, as both Minsky and Papert already knew that multilayer perceptrons were capable of producing an XOR function [9]-

Backpropagation

In 1974, Paul Werbos started to end the AI winter concerning neural networks, when he first described the mathematical process of training multilayer perceptrons through backpropagation of errors [10], derived in the context of control theory by Henry J. Kelley in 1960 [11] and by Arthur E. Bryson in 1961 [12] using principles of dynamic programming, simplified by Stuart E. Dreyfus in 1961 applying the chain rule [13]. It was in 1982, when Werbos applied a automatic differentiation method described in 1970 by Seppo Linnainmaa [14] to neural networks in the way that is widely used today [15] [16] [17] [18].

Backpropagation is a generalization of the delta rule to multilayered feedforward networks, made possible by using the chain rule to iteratively compute gradients for each layer. Backpropagation requires that the activation function used by the artificial neurons be differentiable, which is true for the common sigmoid logistic function or its softmax generalization in multiclass classification.

Along with an optimization method such as gradient descent, it calculates the gradient of a cost or loss function with respect to all the weights in the neural network. The gradient is fed to the optimization method which in turn uses it to update the weights, in an attempt to minimize the loss function, which choice depends on the learning type (supervised, unsupervised, reinforcement) and the activation function - mean squared error or cross-entropy error function are used in binary classification [19]. The gradient is almost always used in a simple stochastic gradient descent algorithm. In 1983, Yurii Nesterov contributed an accelerated version of gradient descent that converges considerably faster than ordinary gradient descent [20] [21] [22] [23].

Backpropagation algorithm for a 3-layer network [24]:

   initialize the weights in the network (often small random values)
   do
      for each example e in the training set
         O = neural-net-output(network, e)  // forward pass
         T = teacher output for e
         compute error (T - O) at the output units
         compute delta_wh for all weights from hidden layer to output layer  // backward pass
         compute delta_wi for all weights from input layer to hidden layer   // backward pass continued
         update the weights in the network
   until all examples classified correctly or stopping criterion satisfied
   return the network

Deep Learning

Deep learning has been characterized as a buzzword, or a rebranding of neural networks. A deep neural network (DNN) is an ANN with multiple hidden layers of units between the input and output layers which can be discriminatively trained with the standard backpropagation algorithm. Two common issues if naively trained are overfitting and computation time.

Convolutional NNs

Convolutional neural networks (CNN) form a subclass of feedforward neural networks that have special weight constraints, individual neurons are tiled in such a way that they respond to overlapping regions. A neuron of a convolutional layer is connected to a correspondent receptive field of the previous layer, a small subset of their neurons. A distinguishing feature of CNNs is that many neurons share the same bias and vector of weights, dubbed filter. This reduces memory footprint because a single bias and a single vector of weights is used across all receptive fields sharing that filter, rather than each receptive field having its own bias and vector of weights. Convolutional NNs are suited for deep learning and are highly suitable for parallelization on GPUs [25]. They were research topic in the game of Go since 2008 [26], and along with the residual modification successful applied in Go and other games, most spectacular due to AlphaGo in 2015 and AlphaZero in 2017.

Typical cnn.png

Typical CNN [27]

Residual Nets

A residual block [28] [29]

Residual nets add the input of a layer, typically composed of a convolutional layer and of a ReLU layer, to its output. This modification, like convolutional nets inspired from image classification, enables faster training and deeper networks [30] [31].

ANNs in Games

Applications of neural networks in computer games and chess are learning of evaluation and search control. Evaluation topics include feature selection and automated tuning, search control move ordering, selectivity and time management. The perceptron looks like the ideal learning algorithm for automated evaluation tuning.

Backgammon

In the late 80s, Gerald Tesauro pioneered in applying ANNs to the game of Backgammon. His program Neurogammon won the Gold medal at the 1st Computer Olympiad 1989 - and was further improved by TD-Lambda based Temporal Difference Learning within TD-Gammon [32]. Today all strong backgammon programs rely on heavily trained neural networks.

Go

In 2014, two teams independently investigated whether deep convolutional neural networks could be used to directly represent and learn a move evaluation function for the game of Go. Christopher Clark and Amos Storkey trained an 8-layer convolutional neural network by supervised learning from a database of human professional games, which without any search, defeated the traditional search program Gnu Go in 86% of the games [33] [34] [35] [36]. In their paper Move Evaluation in Go Using Deep Convolutional Neural Networks [37], Chris J. Maddison, Aja Huang, Ilya Sutskever, and David Silver report they trained a large 12-layer convolutional neural network in a similar way, to beat Gnu Go in 97% of the games, and matched the performance of a state-of-the-art Monte-Carlo Tree Search that simulates a million positions per move [38].

In 2015, a team affiliated with Google DeepMind around David Silver and Aja Huang, supported by Google researchers John Nham and Ilya Sutskever, build a Go playing program dubbed AlphaGo [39], combining Monte-Carlo tree search with their 12-layer networks [40].

Chess

Logistic regression as applied in Texel's Tuning Method may be interpreted as supervised learning application of the single-layer perceptron with one neuron. This is also true for reinforcement learning approaches, such as TD-Leaf in KnightCap or Meep's TreeStrap, where the evaluation consists of a weighted linear combination of features. Despite these similarities with the perceptron, these engines are not considered using ANNs - since they use manually selected chess specific feature construction concepts like material, piece square tables, pawn structure, mobility etc..

More sophisticated attempts to replace static evaluation by neural networks and perceptrons feeding in more unaffiliated feature sets like board representation and attack tables etc., where not yet that successful like in other games. Chess evaluation seems not that well suited for neural nets, but there are also aspects of too weak models and feature recognizers as addressed by Gian-Carlo Pascutto with Stoofvlees [41], huge training effort, and weak floating point performance - but there is still hope due to progress in hardware and parallelization using SIMD instructions and GPUs, and deeper and more powerful neural network structures and methods successful in other domains. In December 2017, Google DeepMind published about their generalized AlphaZero algorithm.

Move Ordering

Concerning move ordering - there were interesting NN proposals like the Chessmaps Heuristic by Kieran Greer et al. [42], and the Neural MoveMap Heuristic by Levente Kocsis et al. [43].

Giraffe & Zurichess

In 2015, Matthew Lai trained Giraffe's deep neural network by TD-Leaf [44]. Zurichess by Alexandru Moșoi uses the TensorFlow library for automated tuning - in a two layers neural network, the second layer is responsible for a tapered eval to phase endgame and middlegame scores [45].

DeepChess

In 2016, Omid E. David, Nathan S. Netanyahu, and Lior Wolf introduced DeepChess obtaining a grandmaster-level chess playing performance using a learning method incorporating two deep neural networks, which are trained using a combination of unsupervised pretraining and supervised training. The unsupervised training extracts high level features from a given chess position, and the supervised training learns to compare two chess positions to select the more favorable one. In order to use DeepChess inside a chess program, a novel version of alpha-beta is used that does not require bounds but positions αpos and βpos [46].

Alpha Zero

In December 2017, the Google DeepMind team along with former Giraffe author Matthew Lai reported on their generalized AlphaZero algorithm, combining Deep learning with Monte-Carlo Tree Search. AlphaZero can achieve, tabula rasa, superhuman performance in many challenging domains with some training effort. Starting from random play, and given no domain knowledge except the game rules, AlphaZero achieved a superhuman level of play in the games of chess and Shogi as well as Go, and convincingly defeated a world-champion program in each case [47].

NN Chess Programs

See also

AlphaGo
Keynote Lecture CG 2016 Conference by Aja Huang

Selected Publications

1940 ...

1950 ...

Claude Shannon, John McCarthy (eds.) (1956). Automata Studies. Annals of Mathematics Studies, No. 34
Claude Shannon, John McCarthy (eds.) (1956). Automata Studies. Annals of Mathematics Studies, No. 34, pdf

1960 ...

1970 ...

1980 ...

1987

1988

1989

1990 ...

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000 ...

2001

2002

2003

2004

2006

2007

2008

2009

  • Daniel Abdi, Simon Levine, Girma T. Bitsuamlak (2009). Application of an Artificial Neural Network Model for Boundary Layer Wind Tunnel Profile Development. 11th Americas conference on wind Engineering, pdf

2010 ...

2011

2012

Nicol N. Schraudolph (2012). Centering Neural Network Gradient Factors.
Léon Bottou (2012). Stochastic Gradient Descent Tricks. Microsoft Research, pdf
Ronan Collobert, Koray Kavukcuoglu, Clément Farabet (2012). Implementing Neural Networks Efficiently. [64]

2013

2014

2015

2016

2017

2018

2019

Blog & Forum Posts

1996 ...

Re: Evaluation by neural network ? by Jay Scott, CCC, November 10, 1997 [85]

2000 ...

Re: Whatever happened to Neural Network Chess programs? by Andy Walker, rgcc, March 28, 2000  » Advances in Computer Chess 1, Ron Atkin
Combining Neural Networks and Alpha-Beta by Matthias Lüscher, rgcc, April 01, 2000 » Chessterfield
Neural nets in backgammon by Albert Silver, CCC, April 07, 2004

2005 ...

2010 ...

Re: Chess program with Artificial Neural Networks (ANN)? by Gian-Carlo Pascutto, CCC, January 07, 2010 » Stoofvlees
Re: Chess program with Artificial Neural Networks (ANN)? by Gian-Carlo Pascutto, CCC, January 08, 2010
Re: Chess program with Artificial Neural Networks (ANN)? by Volker Annuss, CCC, January 08, 2010 » Hermann

2015 ...

2016

Re: Deep Learning Chess Engine ? by Alexandru Mosoi, CCC, July 21, 2016 » Zurichess
Re: Deep Learning Chess Engine ? by Matthew Lai, CCC, August 04, 2016 » Giraffe [89]

2017

Re: Is AlphaGo approach unsuitable to chess? by Peter Österlund, CCC, May 31, 2017 » Texel

2018

2019

Re: A question to MCTS + NN experts by Daniel Shawul, CCC, July 17, 2019

External Links

Biological

ANNs

Topics

Neurogrid from Wikipedia

Perceptron

History of the Perceptron

CNNs

Convolutional Neural Networks
Deep Residual Networks
An Introduction to different Types of Convolutions in Deep Learning by Paul-Louis Pröve, July 22, 2017
Squeeze-and-Excitation Networks by Paul-Louis Pröve, October 17, 2017

RNNs

Restricted Boltzmann machine from Wikipedia

Activation Functions

Backpropagation

Gradient

Momentum from Wikipedia

Software

Neural Lab from Wikipedia
SNNS from Wikipedia

Libraries

Blogs

The Single Layer Perceptron
The Sigmoid Function in C#
Hidden Neurons and Feature Space
Training Neural Networks Using Back Propagation in C#
Data Mining with Artificial Neural Networks (ANN)
Neural Net in C++ Tutorial on Vimeo (also on YouTube)

Courses

Part 1: Data and Architecture, YouTube Videos
Part 2: Forward Propagation
Part 3: Gradient Descent
Part 4: Backpropagation
Part 5: Numerical Gradient Checking
Part 6: Training
Part 7: Overfitting, Testing, and Regularization
But what *is* a Neural Network? | Chapter 1
Gradient descent, how neural networks learn | Chapter 2
What is backpropagation really doing? | Chapter 3
Backpropagation calculus | Appendix to Chapter 3
Lecture 1 | Introduction to Convolutional Neural Networks for Visual Recognition by Justin Johnson, slides
Lecture 2 | Image Classification by Justin Johnson, slides
Lecture 3 | Loss Functions and Optimization by Justin Johnson, slides
Lecture 4 | Introduction to Neural Networks by Serena Yeung, slides
Lecture 5 | Convolutional Neural Networks by Serena Yeung, slides
Lecture 6 | Training Neural Networks I by Serena Yeung, slides
Lecture 7 | Training Neural Networks II by Justin Johnson, slides
Lecture 8 | Deep Learning Software by Justin Johnson, slides
Lecture 9 | CNN Architectures by Serena Yeung, slides
Lecture 10 | Recurrent Neural Networks by Justin Johnson, slides
Lecture 11 | Detection and Segmentation by Justin Johnson, slides
Lecture 12 | Visualizing and Understanding by Justin Johnson, slides
Lecture 13 | Generative Models by Serena Yeung, slides
Lecture 14 | Deep Reinforcement Learning by Serena Yeung, slides
Lecture 15 | Efficient Methods and Hardware for Deep Learning by Song Han, slides

Music

Marc Ribot, Kenny Wollesen, Joey Baron, Jamie Saft, Trevor Dunn, Cyro Baptista, John Zorn

References

  1. An example artificial neural network with a hidden layer, Image by Colin M.L. Burnett with Inkscape, December 27, 2006, CC BY-SA 3.0, Artificial Neural Networks/Neural Network Basics - Wikibooks, Wikimedia Commons
  2. Biological neural network - Early study - from Wikipedia
  3. Warren S. McCulloch, Walter Pitts (1943). A Logical Calculus of the Ideas Immanent in Nervous Activity. Bulletin of Mathematical Biology, Vol. 5, No. 1, pdf
  4. Artificial neuron from Wikipedia
  5. The appropriate weights are applied to the inputs, and the resulting weighted sum passed to a function that produces the output y, image created by mat_the_w, based on raster image Perceptron.gif by 'Paskari', using Inkscape 0.46 for OSX, Wikimedia Commons, Perceptron from Wikipedia
  6. Frank Rosenblatt (1957). The Perceptron - a Perceiving and Recognizing Automaton. Report 85-460-1, Cornell Aeronautical Laboratory
  7. A two-layer neural network capable of calculating XOR. The numbers within the neurons represent each neuron's explicit threshold (which can be factored out so that all neurons have the same threshold, usually 1). The numbers that annotate arrows represent the weight of the inputs. This net assumes that if the threshold is not reached, zero (not -1) is output. Note that the bottom layer of inputs is not always considered a real neural network layer, Feedforward neural network from Wikipedia
  8. multilayer perceptron is a misnomer for a more complicated neural network
  9. Perceptron from Wikipedia
  10. Paul Werbos (1974). Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences. Ph. D. thesis, Harvard University
  11. Henry J. Kelley (1960). Gradient Theory of Optimal Flight Paths. [http://arc.aiaa.org/loi/arsj ARS Journal, Vol. 30, No. 10
  12. Arthur E. Bryson (1961). A gradient method for optimizing multi-stage allocation processes. In Proceedings of the Harvard University Symposium on digital computers and their applications
  13. Stuart E. Dreyfus (1961). The numerical solution of variational problems. RAND paper P-2374
  14. Seppo Linnainmaa (1970). The representation of the cumulative rounding error of an algorithm as a Taylor expansion of the local rounding errors. Master's thesis, University of Helsinki
  15. Paul Werbos (1982). Applications of advances in nonlinear sensitivity analysis. System Modeling and Optimization, Springer, pdf
  16. Paul Werbos (1994). The Roots of Backpropagation. From Ordered Derivatives to Neural Networks and Political Forecasting. John Wiley & Sons
  17. Deep Learning - Scholarpedia | Backpropagation by Jürgen Schmidhuber
  18. Who Invented Backpropagation? by Jürgen Schmidhuber (2014, 2015)
  19. "Using cross-entropy error function instead of sum of squares leads to faster training and improved generalization", from Sargur Srihari, Neural Network Training (pdf)
  20. Yurii Nesterov from Wikipedia
  21. ORF523: Nesterov’s Accelerated Gradient Descent by Sébastien Bubeck, I’m a bandit, April 1, 2013
  22. Nesterov’s Accelerated Gradient Descent for Smooth and Strongly Convex Optimization by Sébastien Bubeck, I’m a bandit, March 6, 2014
  23. Revisiting Nesterov’s Acceleration by Sébastien Bubeck, I’m a bandit, June 30, 2015
  24. Backpropagation algorithm from Wikipedia
  25. PARsE | Education | GPU Cluster | Efficient mapping of the training of Convolutional Neural Networks to a CUDA-based cluster
  26. Ilya Sutskever, Vinod Nair (2008). Mimicking Go Experts with Convolutional Neural Networks. ICANN 2008, pdf
  27. Typical CNN architecture, Image by Aphex34, December 16, 2015, CC BY-SA 4.0, Wikimedia Commons
  28. The fundamental building block of residual networks. Figure 2 in Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun (2015). Deep Residual Learning for Image Recognition. arXiv:1512.03385
  29. Understand Deep Residual Networks — a simple, modular learning framework that has redefined state-of-the-art by Michael Dietz, Waya.ai, May 02, 2017
  30. Tristan Cazenave (2017). Residual Networks for Computer Go. IEEE Transactions on Computational Intelligence and AI in Games, Vol. PP, No. 99, pdf
  31. Deep Residual Networks from TUM Wiki, Technical University of Munich
  32. Richard Sutton, Andrew Barto (1998). Reinforcement Learning: An Introduction. MIT Press, 11.1 TD-Gammon
  33. Christopher Clark, Amos Storkey (2014). Teaching Deep Convolutional Neural Networks to Play Go. arXiv:1412.3409
  34. Teaching Deep Convolutional Neural Networks to Play Go by Hiroshi Yamashita, The Computer-go Archives, December 14, 2014
  35. Why Neural Networks Look Set to Thrash the Best Human Go Players for the First Time | MIT Technology Review, December 15, 2014
  36. Teaching Deep Convolutional Neural Networks to Play Go by Michel Van den Bergh, CCC, December 16, 2014
  37. Chris J. Maddison, Aja Huang, Ilya Sutskever, David Silver (2014). Move Evaluation in Go Using Deep Convolutional Neural Networks. arXiv:1412.6564v1
  38. Move Evaluation in Go Using Deep Convolutional Neural Networks by Aja Huang, The Computer-go Archives, December 19, 2014
  39. AlphaGo | Google DeepMind
  40. David Silver, Aja Huang, Chris J. Maddison, Arthur Guez, Laurent Sifre, George van den Driessche, Julian Schrittwieser, Ioannis Antonoglou, Veda Panneershelvam, Marc Lanctot, Sander Dieleman, Dominik Grewe, John Nham, Nal Kalchbrenner, Ilya Sutskever, Timothy Lillicrap, Madeleine Leach, Koray Kavukcuoglu, Thore Graepel, Demis Hassabis (2016). Mastering the game of Go with deep neural networks and tree search. Nature, Vol. 529
  41. Re: Chess program with Artificial Neural Networks (ANN)? by Gian-Carlo Pascutto, CCC, January 07, 2010
  42. Kieran Greer, Piyush Ojha, David A. Bell (1999). A Pattern-Oriented Approach to Move Ordering: the Chessmaps Heuristic. ICCA Journal, Vol. 22, No. 1
  43. Levente Kocsis, Jos Uiterwijk, Eric Postma, Jaap van den Herik (2002). The Neural MoveMap Heuristic in Chess. CG 2002
  44. *First release* Giraffe, a new engine based on deep learning by Matthew Lai, CCC, July 08, 2015
  45. Re: Deep Learning Chess Engine ? by Alexandru Mosoi, CCC, July 21, 2016
  46. Omid E. David, Nathan S. Netanyahu, Lior Wolf (2016). DeepChess: End-to-End Deep Neural Network for Automatic Learning in Chess. ICAAN 2016, Lecture Notes in Computer Science, Vol. 9887, Springer, pdf preprint
  47. David Silver, Thomas Hubert, Julian Schrittwieser, Ioannis Antonoglou, Matthew Lai, Arthur Guez, Marc Lanctot, Laurent Sifre, Dharshan Kumaran, Thore Graepel, Timothy Lillicrap, Karen Simonyan, Demis Hassabis (2017). Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm. arXiv:1712.01815
  48. Rosenblatt's Contributions
  49. The abandonment of connectionism in 1969 - Wikipedia
  50. Frank Rosenblatt (1962). Principles of Neurodynamics: Perceptrons and the Theory of Brain Mechanisms. Spartan Books
  51. Seppo Linnainmaa (1976). Taylor expansion of the accumulated rounding error. BIT Numerical Mathematics, Vol. 16, No. 2
  52. Backpropagation from Wikipedia
  53. Paul Werbos (1994). The Roots of Backpropagation. From Ordered Derivatives to Neural Networks and Political Forecasting. John Wiley & Sons
  54. Neocognitron - Scholarpedia by Kunihiko Fukushima
  55. Classical conditioning from Wikipedia
  56. Sepp Hochreiter's Fundamental Deep Learning Problem (1991) by Jürgen Schmidhuber, 2013
  57. Nici Schraudolph’s go networks, review by Jay Scott
  58. Re: Evaluation by neural network ? by Jay Scott, CCC, November 10, 1997
  59. Long short term memory from Wikipedia
  60. Tsumego from Wikipedia
  61. Helmholtz machine from Wikipedia
  62. Who introduced the term “deep learning” to the field of Machine Learning by Jürgen Schmidhuber, Google+, March 18, 2015
  63. Presentation for a neural net learning chess program by Dann Corbit, CCC, April 06, 2004
  64. Clément Farabet | Code
  65. Demystifying Deep Reinforcement Learning by Tambet Matiisen, Nervana, December 21, 2015
  66. Generative adversarial networks from Wikipedia
  67. Teaching Deep Convolutional Neural Networks to Play Go by Hiroshi Yamashita, The Computer-go Archives, December 14, 2014
  68. Teaching Deep Convolutional Neural Networks to Play Go by Michel Van den Bergh, CCC, December 16, 2014
  69. Arasan 19.2 by Jon Dart, CCC, November 03, 2016 » Arasan's Tuning
  70. GitHub - BarakOshri/ConvChess: Predicting Moves in Chess Using Convolutional Neural Networks
  71. ConvChess CNN by Brian Richardson, CCC, March 15, 2017
  72. Jürgen Schmidhuber (2015) Critique of Paper by "Deep Learning Conspiracy" (Nature 521 p 436).
  73. How Facebook’s AI Researchers Built a Game-Changing Go Engine | MIT Technology Review, December 04, 2015
  74. Combining Neural Networks and Search techniques (GO) by Michael Babigian, CCC, December 08, 2015
  75. DeepChess: Another deep-learning based chess program by Matthew Lai, CCC, October 17, 2016
  76. ICANN 2016 | Recipients of the best paper awards
  77. Jigsaw puzzle from Wikipedia
  78. CMA-ES from Wikipedia
  79. catastrophic forgetting by Daniel Shawul, CCC, May 09, 2019
  80. Using GAN to play chess by Evgeniy Zheltonozhskiy, CCC, February 23, 2017
  81. AlphaGo Zero: Learning from scratch by Demis Hassabis and David Silver, DeepMind, October 18, 2017
  82. Google's AlphaGo team has been working on chess by Peter Kappler, CCC, December 06, 2017
  83. Residual Networks for Computer Go by Brahim Hamadicharef, CCC, December 07, 2017
  84. AlphaZero: Shedding new light on the grand games of chess, shogi and Go by David Silver, Thomas Hubert, Julian Schrittwieser and Demis Hassabis, DeepMind, December 03, 2018
  85. Alois Heinz (1994). Efficient Neural Net α-β-Evaluators. pdf
  86. Mathieu Autonès, Aryel Beck, Phillippe Camacho, Nicolas Lassabe, Hervé Luga, François Scharffe (2004). Evaluation of Chess Position by Modular Neural network Generated by Genetic Algorithm. EuroGP 2004
  87. Naive Bayes classifier from Wikipedia
  88. GitHub - pluskid/Mocha.jl: Deep Learning framework for Julia
  89. Rectifier (neural networks) from Wikipedia
  90. Muthuraman Chidambaram, Yanjun Qi (2017). Style Transfer Generative Adversarial Networks: Learning to Play Chess Differently. arXiv:1702.06762v1
  91. erikbern/deep-pink · GitHub
  92. Neural networks (NN) explained by Erin Dame, CCC, December 20, 2017

Up one Level