Changes

Jaime Carbonell invented multiple well-known algorithms and methods, including [https://en.wikipedia.org/wiki/Proactivity proactive] [[Learning|machine learning]] for multi-source cost-sensitive [https://en.wikipedia.org/wiki/Active_learning active learning], linked [https://en.wikipedia.org/wiki/Conditional_random_field conditional random fields] (L-SCRF) for predicting tertiary and quaternary [https://en.wikipedia.org/wiki/Protein_folding protein folds], [https://en.wikipedia.org/wiki/Automatic_summarization maximal marginal relevance] (MMR) for information novelty, retrieval and summarization, topic-conditioned modeling for [https://en.wikipedia.org/wiki/Novelty_detection novelty detection], symmetric optimal [https://en.wikipedia.org/wiki/Phrase phrasal] alignment method for trainable example-based and [https://en.wikipedia.org/wiki/Statistical_machine_translation statistical machine translation], series-anomaly modeling for [https://en.wikipedia.org/wiki/Predictive_analytics#Fraud_detection financial fraud detection] and [https://en.wikipedia.org/wiki/Clinical_surveillance syndromic surveillance], knowledge-based[https://en.wikipedia.org/wiki/Interlingual_machine_translation interlingual machine translation], transformational analogy for [https://en.wikipedia.org/wiki/Case-based_reasoning case-based reasoning], [https://en.wikipedia.org/wiki/Derivation_%28linguistics%29 derivational] analogy for reconstructive [https://en.wikipedia.org/wiki/Theory_of_justification justification-based] reasoning, robust case-frame parsing, seeded [https://en.wikipedia.org/wiki/Version_space version-space learning], and developed improvements to several other machine learning algorithms. Current research foci include robust statistical learning and mapping [https://en.wikipedia.org/wiki/Peptide_sequence protein sequences] to 3D structure and inferring functional properties, automated transfer-rule learning for machine translation, enriched active transfer learning context-based machine translation, and machine translation for very rare languages <ref>[http://www.cs.cmu.edu/~jgc/CVs/Carbonell%20CV%202013-March.pdf Jaime G. Carbonell - Curriculum Vita] (pdf)</ref>.

* [[Adams Wei Yu]], [[Lei Huang]], [[Qihang Lin]], [[Mathematician#RRSalakhutdinov|Ruslan Salakhutdinov]], [[Jaime Carbonell]] ('''2017'''). ''Block-Normalized Gradient Method: An Empirical Study for Training Deep Neural Network''. [https://arxiv.org/abs/1707.04822 arXiv:1707.04822]