conftrace_
← Authors

Philip Thomas

17 papers · 2014–2023 · 3 conferences · across top CS/AI conferences

Achievements

Jump to papers ↓
+9 more ↓ 🧭 Keyword Pioneer 🐣 Hot Topic Early Bird πŸŒ‰ Interdisciplinary Bridge πŸ—ΊοΈ Taxonomy Completionist (15) 🌍 Conference Polyglot (3)
πŸ—ΊοΈ Taxonomy Completionist (15) 🧭 Keyword Pioneer 🐣 Hot Topic Early Bird 🧬 Topic Evolution πŸ† Keyword Champion (2) πŸ—ƒοΈ Keyword Collector (78) πŸ“ˆ Trend Setter πŸ’Ž Century Club (17) ⚑ Prolific Year (5)

Conferences

ICML (14) AAAI (2) AISTATS (1)

Top co-authors

Yash Chandak (7) Georgios Theocharous (6) Emma Brunskill (3) Chris Nota (3) James Kostas (3) Erik Learned-Miller (2) Christoph Dann (2) Scott Jordan (2) Martha White (2) Blossom Metevier (1)

Keywords

reinforcement learning (6) policy gradient (5) markov decision process (2) neural network (2) fisher information matrix (2) variance reduction (2) action space (2) convergence guarantee (2) statistical learning (2) doubly robust estimator (2) gradient descent (2) dynamic regret (1) risk management (1) hierarchical reinforcement learning (1) evaluation methodology (1) policy learning (1) off-policy evaluation (1) representation learning (1) convergence analysis (1) sequential decision making (1)

Papers

Asymptotically Unbiased Off-Policy Policy Evaluation when Reusing Old Data in Nonstationary Environments AISTATS 2023 Posterior Value Functions: Hindsight Baselines for Policy Gradient Methods ICML 2021 High Confidence Generalization for Reinforcement Learning ICML 2021 Towards Practical Mean Bounds for Small Samples ICML 2021 Lifelong Learning with a Changing Action Set AAAI 2020 Reinforcement Learning When All Actions Are Not Always Available AAAI 2020 Optimizing for the Future in Non-Stationary MDPs ICML 2020 Evaluating the Performance of Reinforcement Learning Algorithms ICML 2020 Asynchronous Coagent Networks ICML 2020 Learning Action Representations for Reinforcement Learning ICML 2019 Concentration Inequalities for Conditional Value at Risk ICML 2019 Decoupling Gradient-Like Learning Rules from Representations ICML 2018 Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learning ICML 2016 Energetic Natural Gradient Descent ICML 2016 High Confidence Policy Improvement ICML 2015 GeNGA: A Generalization of Natural Gradient Ascent with Positive and Negative Convergence Results ICML 2014 Bias in Natural Actor-Critic Algorithms ICML 2014