Amos Storkey

From Chessprogramming wiki
Jump to: navigation, search

Home * People * Amos Storkey

Amos Storkey [1]

Amos James Storkey,
a British computer scientist and reader (associate professor) at Institute for Adaptive and Neural Computation, School of Informatics, University of Edinburgh. He holds a Ph.D. in 1999 on neural networks from the Neural Systems Group, Department of Electrical Engineering, Imperial College London under Philippe De Wilde [2]. His research covers machine learning applied to prediction market, astronomy and changing environments, bayesian modeling in neuroimaging, learning and inference, dynamical Boltzmann machine models, and scalable deep learning [3].

DCNNs in Go

As reported in their 2014 paper Teaching Deep Convolutional Neural Networks to Play Go, Amos Storkey along with Christopher Clark trained an 8-layer convolutional neural network [4] by supervised learning from a database of human professional games to predict the moves made by expert Go players. They introduced a number of novel techniques, including a method of tying weights in the network to 'hard code' symmetries that are expect to exist in the target function, and demonstrated in an ablation study they considerably improve performance. Their final networks can consistently defeat Gnu Go , indicating it is state of the art among programs that do not use Monte-Carlo Tree Search, and was also able to win some games against Fuego while using a fraction of the playing time [5] [6] [7] [8] [9] .

Selected Publications

[10]

1997 ...

2000 ...

2010 ...

External Links

References

Up one level