Scholar

George Tucker

Google Scholar ID: -gJkPHIAAAAJ

Google DeepMind

Reinforcement Learning

Citations & Impact

All-time

Citations

35,354

H-index

i10-index

Publications

Co-authors

list available

Contact

Publications

1 items

Cited

Resume

Academic Achievements

Published multiple influential papers on offline reinforcement learning in top-tier conferences including NeurIPS, ICLR, and ICML
Notable works include: 'Conservative Q-Learning for Offline Reinforcement Learning' (NeurIPS 2020), 'Offline reinforcement learning: Tutorial, review, and perspectives on open problems'
Paper 'Offline Q-Learning on Diverse Multi-Task Data Both Scales And Generalizes' accepted at ICLR 2023 (Top 5%)
Contributions in gradient estimation, variational bounds of mutual information, and Bayesian deep networks, e.g., 'DisARM: An antithetic gradient estimator for binary latent variables' (NeurIPS 2020 Spotlight)
Developed and open-sourced projects such as DReGS (sites.google.com/corp/view/dregs) and Mirage-RL (github.com/brain-research/mirage-rl)

Co-authors

15 total