Philip S. Thomas

发表

Safe Reinforcement Learning

Philip S. Thomas, Philip Thomas, 2015 .

Importance Sampling for Fair Policy Selection

Philip S. Thomas, Emma Brunskill, Shayan Doroudi, 2017, UAI.

Security Analysis of Safe and Seldonian Reinforcement Learning Algorithms

Philip S. Thomas, Pinar Ozisik, P. Thomas, 2020, NeurIPS.

Asynchronous Coagent Networks: Stochastic Networks for Reinforcement Learning without Backpropagation or a Clock pdf

Philip S. Thomas, James Kostas, Chris Nota, 2019, ArXiv.

Learning Reusable Options for Multi-Task Reinforcement Learning pdf

Philip S. Thomas, Chris Nota, Francisco M. Garcia, 2020, ArXiv.

Conjugate Markov Decision Processes

Andrew G. Barto, Philip S. Thomas, A. Barto, 2011, ICML.

Decoupling Learning Rules from Representations pdf

Philip S. Thomas, Christoph Dann, Emma Brunskill, 2017, ArXiv.

Proximal Reinforcement Learning: A New Theory of Sequential Decision Making in Primal-Dual Spaces pdf

Bo Liu, Ji Liu, Sridhar Mahadevan, 2014, ArXiv.

Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learning pdf

Philip S. Thomas, Emma Brunskill, P. Thomas, 2016, ICML.

Motor primitive discovery

Andrew G. Barto, Philip S. Thomas, A. Barto, 2012, 2012 IEEE International Conference on Development and Learning and Epigenetic Robotics (ICDL).

Using Options for Long-Horizon Off-Policy Evaluation pdf

Philip S. Thomas, Emma Brunskill, Zhaohan Daniel Guo, 2017, ArXiv.

Data-Efficient Policy Evaluation Through Behavior Policy Search

Peter Stone, Scott Niekum, Philip S. Thomas, 2017, ICML.

Asynchronous Coagent Networks: Stochastic Networks for Reinforcement Learning without Backpropagation or a Clock

Philip S. Thomas, Chris Nota, James E. Kostas, 2019, ICML.

Is the Policy Gradient a Gradient? pdf

Philip S. Thomas, Chris Nota, P. Thomas, 2019, AAMAS.

Ad Recommendation Systems for Life-Time Value Optimization

Philip S. Thomas, Mohammad Ghavamzadeh, Georgios Theocharous, 2015, WWW.

A Compression-Inspired Framework for Macro Discovery

Philip S. Thomas, Francisco M. Garcia, Bruno C. da Silva, 2017, AAMAS.

High-Confidence Off-Policy Evaluation

Philip S. Thomas, Mohammad Ghavamzadeh, Georgios Theocharous, 2015, AAAI.

Policy Gradient Methods for Reinforcement Learning with Function Approximation and Action-Dependent Baselines pdf

Philip S. Thomas, Emma Brunskill, P. Thomas, 2017, ArXiv.

Application of the Actor-Critic Architecture to Functional Electrical Stimulation Control of a Human Arm

Philip S. Thomas, Kathleen M. Jagodnik, Antonie J. van den Bogert, 2009, IAAI.

Energetic Natural Gradient Descent

Bruno Castro da Silva, Philip S. Thomas, Christoph Dann, 2016, ICML.

Natural Temporal Difference Learning

Philip S. Thomas, William Dabney, P. Thomas, 2014, AAAI.

Double Loop Process: The Next Innovation Paradigm

Philip S. Thomas, 2003 .

Using Options and Covariance Testing for Long Horizon Off-Policy Policy Evaluation

Philip S. Thomas, Emma Brunskill, Zhaohan Guo, 2017, NIPS.

Decoupling Gradient-Like Learning Rules from Representations

Philip S. Thomas, Christoph Dann, Emma Brunskill, 2018, ICML.

Human-Like Rewards to Train a Reinforcement Learning Controller for Planar Arm Movement

Philip S. Thomas, Kathleen M. Jagodnik, Antonie J. van den Bogert, 2016, IEEE Transactions on Human-Machine Systems.

Evaluating the Performance of Reinforcement Learning Algorithms pdf

Daniel Cohen, Philip S. Thomas, Mengxue Zhang, 2020, ICML.

Offline Contextual Bandits with High Probability Fairness Guarantees

Yuriy Brun, Ari Kobren, Philip S. Thomas, 2019, NeurIPS.

Importance Sampling with Unequal Support pdf

Philip S. Thomas, Emma Brunskill, P. Thomas, 2016, AAAI.

A Notation for Markov Decision Processes pdf

Philip S. Thomas, P. Thomas, 2015, ArXiv.

TD_gamma: Re-evaluating Complex Backups in Temporal Difference Learning

Scott Niekum, George Konidaris, Philip S. Thomas, 2011, NIPS.

High-Confidence Off-Policy (or Counterfactual) Variance Estimation

Philip S. Thomas, Shiv Shankar, Yash Chandak, 2021, AAAI.

Training an Actor-Critic Reinforcement Learning Controller for Arm Movement Using Human-Generated Rewards

Philip S. Thomas, Kathleen M. Jagodnik, Robert F. Kirsch, 2017, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

Increasing the Action Gap: New Operators for Reinforcement Learning

Marc G. Bellemare, Rémi Munos, Philip S. Thomas, 2015, AAAI.

Learning Action Representations for Reinforcement Learning pdf

Philip S. Thomas, Georgios Theocharous, James Kostas, 2019, ICML.

Projected Natural Actor-Critic

Sridhar Mahadevan, Philip S. Thomas, Stephen Giguere, 2013, NIPS.

Classical Policy Gradient: Preserving Bellman's Principle of Optimality pdf

Philip S. Thomas, James Kostas, Chris Nota, 2019, ArXiv.

Reinforcement Learning Without Backpropagation or a Clock

Philip S. Thomas, James Kostas, Chris Nota, 2019 .

Privacy Preserving Off-Policy Evaluation pdf

Philip S. Thomas, Gerome Miklau, Tengyang Xie, 2019, ArXiv.

Value Function Approximation in Reinforcement Learning Using the Fourier Basis

George Konidaris, Philip S. Thomas, Sarah Osentoski, 2011, AAAI.

Predictive Off-Policy Policy Evaluation for Nonstationary Decision Problems, with Applications to Digital Marketing

Philip S. Thomas, Mohammad Ghavamzadeh, Georgios Theocharous, 2017, AAAI.

High Confidence Policy Improvement

Philip S. Thomas, Mohammad Ghavamzadeh, Georgios Theocharous, 2015, ICML.

Reinforcement learning with a network of spiking agents

Hananel Hazan, Robert Kozma, Philip S. Thomas, 2019, 1910.06489.

Natural Option Critic pdf

Philip S. Thomas, Saket Tiwari, 2019, AAAI.

Concentration Inequalities for Conditional Value at Risk

Philip S. Thomas, Erik G. Learned-Miller, P. Thomas, 2019, ICML.

Policy Gradient Coagent Networks

Philip S. Thomas, P. Thomas, 2011, NIPS.

A Meta-MDP Approach to Exploration for Lifelong Reinforcement Learning

Philip S. Thomas, Francisco M. Garcia, P. Thomas, 2019, AAMAS.

Optimizing for the Future in Non-Stationary MDPs pdf

Sridhar Mahadevan, Philip S. Thomas, Georgios Theocharous, 2020, ICML.

Reinforcement Learning for Strategic Recommendations pdf

Philip S. Thomas, Frits de Nijs, Georgios Theocharous, 2020, ArXiv.

Reinforcement Learning When All Actions are Not Always Available pdf

Philip S. Thomas, Georgios Theocharous, Yash Chandak, 2019, AAAI.

Towards Safe Policy Improvement for Non-Stationary MDPs pdf

Philip S. Thomas, Georgios Theocharous, Martha White, 2020, NeurIPS.

Reinforcement learning with spiking coagents pdf

Hananel Hazan, Robert Kozma, Philip S. Thomas, 2019, ArXiv.

Policy Evaluation Using the Ω-Return

Scott Niekum, George Konidaris, Philip S. Thomas, 2015, NIPS.

On Ensuring that Intelligent Machines Are Well-Behaved pdf

Bruno Castro da Silva, Andrew G. Barto, Philip S. Thomas, 2017, ArXiv.

Personalized Ad Recommendation Systems for Life-Time Value Optimization with Guarantees

Philip S. Thomas, Mohammad Ghavamzadeh, Georgios Theocharous, 2015, IJCAI.

Lifelong Learning with a Changing Action Set pdf

Philip S. Thomas, Georgios Theocharous, Chris Nota, 2019, AAAI.

A New Confidence Interval for the Mean of a Bounded Random Variable pdf

Philip S. Thomas, Erik G. Learned-Miller, P. Thomas, 2019, ArXiv.

Towards Practical Mean Bounds for Small Samples

Philip S. Thomas, Erik Learned-Miller, My Phan, 2021, ICML.

High Confidence Generalization for Reinforcement Learning

Philip S. Thomas, Georgios Theocharous, Scott M. Jordan, 2021, ICML.

SOPE: Spectrum of Off-Policy Estimators pdf

Scott Niekum, Philip S. Thomas, Stephen Giguere, 2021, NeurIPS.

Universal Off-Policy Evaluation pdf

Bruno Castro da Silva, Scott Niekum, Philip S. Thomas, 2021, NeurIPS.

Large-scale Interactive Conversational Recommendation System using Actor-Critic Framework

James Allan, Philip S. Thomas, Ali Montazeralghaem, 2021, RecSys.

Edge-Compatible Reinforcement Learning for Recommendations pdf

Philip S. Thomas, Georgios Theocharous, James E. Kostas, 2021, ArXiv.

Multi-Objective SPIBB: Seldonian Offline Policy Improvement with Safety Constraints in Finite MDPs pdf

Romain Laroche, Joelle Pineau, Philip S. Thomas, 2021, NeurIPS.

Magical Policy Search : Data Efficient Reinforcement Learning with Guarantees of Global Optimality

Philip S. Thomas, 2016 .

Policy Evaluation Using the \Omega -Return

Scott Niekum, George Konidaris, Philip S. Thomas, 2015, NIPS 2015.

Optimizing for the Future in Non-Stationary MDPs pdf

Martha White, Sridhar Mahadevan, Philip S. Thomas, 2020, ICML.