Scholar

R Devon Hjelm

Google Scholar ID: 68c5HfwAAAAJ

Apple MLR, Mila

deep learningrepresentation learningself-supervised learninggenerative models

Citations & Impact

All-time

Citations

14,747

H-index

i10-index

Publications

Co-authors

125

list available

Contact

Publications

20 items

Browse publications on Google Scholar (top-right) ↗

Resume (English only)

Academic Achievements

2019 ICLR oral presentation (top 1.5%): 'Learning deep representations by mutual information estimation and maximization' (a.k.a. Deep InfoMax/DIM)
2019 ICLR conference paper: 'Deep Graph InfoMax'
2019 AAAI paper: 'Online Adaptive Curriculum Learning for GANs'
Two workshop papers accepted at NeurIPS 2018, one on representation learning
2018 ICLR paper: 'Boundary Seeking GANs'
2018 ICML paper: 'Mutual Information Neural Estimation'
Multiple publications in top-tier conferences including NeurIPS, ICML, ICLR, and AAAI

Background

Deep Learning Researcher at Microsoft Research (MSR) Montreal
Adjunct Professor at the University of Montreal
Associate member of Mila
Research focuses on mutual information estimation and self-supervision in representation learning for applications in computer vision, NLP, and RL
Long-term goal: using machine learning to learn representations that aid scientific discovery

Co-authors

125 total