Published multiple papers on topics such as benchmarking neural network training algorithms, measuring the effects of data parallelism on neural network training, and the importance of generation order in language modeling.
Research Experience
During his PhD, collaborated to train the first successful deep acoustic models for automatic speech recognition and led the team that won the Merck molecular activity challenge on Kaggle.
Education
PhD from the University of Toronto, Department of Computer Science, Machine Learning Group, supervised by Geoffrey Hinton.
Background
Research interests include deep learning, natural language processing, and most of statistical machine learning. Currently a research scientist at Google on the Brain team in Mountain View.