Scholar
Emmanuel Bengio
Google Scholar ID: yVtSOt8AAAAJ
McGill University; Recursion/Valence Labs
Citations & Impact
Citations
5,029
H-index
21
i10-index
30
Publications
20
Co-authors
2
list available
Contact
Publications
Resume (English only)
Academic Achievements
- 2021: Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation; 2020: Correcting Momentum in Temporal Difference Learning; Interference and Generalization in Temporal Difference Learning; 2019: Attack and Defense in Cellular Decision-Making: Lessons from Machine Learning; 2018: Disentangling the independently controllable factors of variation by interacting with the world; 2017: World Knowledge for Reading Comprehension: Rare Entity Prediction with Hierarchical LSTMs Using External Descriptions; A Closer Look at Memorization in Deep Networks; Independently Controllable Features; Deep Nets Don't Learn via Memorization; 2016: On Reinforcement Learning for Deep Neural Architectures: conditional computation with stochastic computation policies; Conditional Computation in Neural Networks for faster models; 2015: Conditional computation in neural networks using a decision-theoretic approach; 2013: Combining modality specific deep neural networks for emotion recognition in video.
Research Experience
- Staff ML Scientist at Valence Labs, Recursion, working on GFlowNet and drug-discovery.
Education
- PhD and CS Master's from McGill, as part of Mila, under the supervision of Joelle Pineau and Doina Precup; Computer Science B.Sc. from DIRO at Université de Montréal.
Background
- Research interests: Machine Learning, especially Deep Learning and Reinforcement Learning and mixing both; compiler design and implementation; programming. Lately, working at the intersection of ML and drug design using the GFlowNet framework.
Miscellany
- Personal interests include writing (on topics such as weight initialization of deep networks, happiness, meaning, PhD, meditation, etc.); Contact email: bengioe@gmail.com
