Jason Weston

---

Research Scientist at Google, NY, USA.
Email:

Call for papers -- Special Issue on Learning Semantics in Machine Learning

---

Current Research & Interests

• Machine Learning
• Fast learning.
  • Label Embedding Trees for Large Multi-Class Tasks published in NIPS 2010.
• Deep learning and using unlabeled data.
• Deep Learning from Temporal Coherence in Video, published in ICML '09
• Deep Learning via Semi-Supervised Embedding, published in ICML '08
• Audio
• Large Scale Music Recommendation.
• Large-Scale Music Annotation and Retrieval: Learning to Rank in Joint Semantic Spaces. Journal of New Music Research.
• Images
• Large Scale Image Annotation.
• Wsabie: Scaling Up To Large Vocabulary Image Annotation. IJCAI, 2011.
• Text
• NLP (almost) from scratch by learning feature representations for text.
• Natural Language Processing (almost) from Scratch, JMLR.
• Using knowledge bases in text embedding (feature learning) models.
• Learning Structured Embeddings of Knowledge Bases, AAAI '11.
• Learning language in a situated environment.
• Towards Understanding Situated Natural Language. AISTATS '10
• Semantic search.
• Learning to Rank with (a Lot of) Word Features, Journal of Information Retrieval.
• Bioinformatics
• Mass spectrometry peptide identification.
• Q-Ranker algorithm published in Journal of Proteome Research.
• Percolator algorithm published in Nature Methods.
• Protein classification and ranking.
• SVM-Fold - Protein Fold and Superfamily prediction by SVM.
• RankProp stand-alone protein ranking webserver. Rankprop also features on the UCSC Human Genome Browser.
• Software
• Torch 5 - Free, Fast Scripting Language-based Machine Learning Library (Lua+C).
• SENNA - Semantic Extraction using a Neural Net Architecture
• Large Scale Transductive SVMs .
• la SVM   online SVM solver
• Spider Machine Learning Toolbox (Matlab).

---