Optimal Policy

Optimal policy research focuses on finding the best course of action within a given system, often modeled as a Markov Decision Process (MDP), to maximize a desired outcome (e.g., reward, efficiency). Current research emphasizes developing efficient algorithms, such as policy gradient methods and diffusion models, to solve these problems, particularly in complex settings with high dimensionality or uncertainty, often incorporating techniques like variance reduction and bias correction. These advancements are significant for various fields, including robotics, finance, and AI, enabling improved decision-making in scenarios ranging from controlling robots to optimizing resource allocation. The development of more efficient and robust algorithms for finding optimal policies continues to be a central focus.

Papers

June 20, 2023

The Unintended Consequences of Discount Regularization: Improving Regularization in Certainty Equivalence Reinforcement Learning
Sarah Rathnam, Sonali Parbhoo, Weiwei Pan, Susan A. Murphy, Finale Doshi-Velez
Optimal Policy State Action Pair Certainty Equivalent Regularization Penalty

June 14, 2023

Theoretical Hardness and Tractability of POMDPs in RL with Partial Online State Information
Ming Shi, Yingbin Liang, Ness Shroff
Markov Decision Process Optimal Policy Lower Bound Regret Analysis Hardness Result Human Like RL Moral Tractability State Observation

June 5, 2023

Bayesian Learning of Optimal Policies in Markov Decision Processes with Countably Infinite State-Space
Saghar Adler, Vijay Subramanian
Markov Decision Process Optimal Policy State Space Markov Chain Discrete Time Bayesian Learning

June 1, 2023

Learning Prescriptive ReLU Networks
Wei Sun, Asterios Tsiourvas
Optimal Policy ReLU Layer Prescriptive Analytics

May 31, 2023

ROSARL: Reward-Only Safe Reinforcement Learning
Geraud Nangue Tasse, Tamlin Love, Mark Nemecek, Steven James, Benjamin Rosman
Reinforcement Learning High Dimensional Optimal Policy Safe Reinforcement Learning Model Free Algorithm Minimax Concave Penalty

May 29, 2023

Experience Filter: Using Past Experiences on Unseen Tasks or Environments
Anil Yildiz, Esen Yel, Anthony L. Corso, Kyle H. Wray, Stefan J. Witwicki, Mykel J. Kochenderfer
Autonomous Vehicle Optimal Policy Environment Feature eXperience Report Observable Markov Decision Process Unseen Environment C POMDPs Unseen Task

May 24, 2023

Regret-Optimal Model-Free Reinforcement Learning for Discounted MDPs with Short Burn-In Time
Xiang Ji, Gen Li
Reinforcement Learning Markov Decision Process Optimal Policy Short Term

May 18, 2023

May 16, 2023

Optimizing pre-scheduled, intermittently-observed MDPs
Patrick Zhong, Federico Rossi, Dylan A. Shell
Optimal Policy Multi Objective Optimization Problem Joint Optimization Robotics Problem Implicit Coordination

May 9, 2023

Distributional Multi-Objective Decision Making
Willem Röpke, Conor F. Hayes, Patrick Mannion, Enda Howley, Ann Nowé, Diederik M. Roijers
Optimal Policy Pareto Front Distributional Approach

May 2, 2023

Sample Efficient Model-free Reinforcement Learning from LTL Specifications with Optimality Guarantees
Daqian Shao, Marta Kwiatkowska
Markov Decision Process Optimal Policy Temporal Logic Model Free Reinforcement Learning Optimality Guarantee

April 24, 2023

Model-Free Learning and Optimal Policy Design in Multi-Agent MDPs Under Probabilistic Agent Dropout
Carmel Fiscko, Soummya Kar, Bruno Sinopoli
Multi Agent Optimal Policy Structured Dropout Robust Policy Free Learning Dropout Method Robust Markov Decision Process Robust Markov Decision

April 22, 2023

Reinforcement Learning with an Abrupt Model Change
Wuxia Chen, Taposh Banerjee, Jemin George, Carl Busart
Reinforcement Learning Optimal Policy Reward Function Optimal Timing Model Shift Optimality Criterion

Optimal Policy

Papers

The Unintended Consequences of Discount Regularization: Improving Regularization in Certainty Equivalence Reinforcement Learning

Theoretical Hardness and Tractability of POMDPs in RL with Partial Online State Information

Stochastic Principal-Agent Problems: Efficient Computation and Learning

Fair and Robust Estimation of Heterogeneous Treatment Effects for Policy Learning

State Regularized Policy Optimization on Data with Dynamics Shift

Bayesian Learning of Optimal Policies in Markov Decision Processes with Countably Infinite State-Space

Learning Prescriptive ReLU Networks

ROSARL: Reward-Only Safe Reinforcement Learning

Experience Filter: Using Past Experiences on Unseen Tasks or Environments

Regret-Optimal Model-Free Reinforcement Learning for Discounted MDPs with Short Burn-In Time

Offline Primal-Dual Reinforcement Learning for Linear MDPs

Learning Optimal Policy for Simultaneous Machine Translation via Binary Search

Offline Reinforcement Learning with Additional Covering Distributions

Bayesian Risk-Averse Q-Learning with Streaming Observations

On the Statistical Efficiency of Mean-Field Reinforcement Learning with General Function Approximation

Optimizing pre-scheduled, intermittently-observed MDPs

Distributional Multi-Objective Decision Making

Sample Efficient Model-free Reinforcement Learning from LTL Specifications with Optimality Guarantees

Model-Free Learning and Optimal Policy Design in Multi-Agent MDPs Under Probabilistic Agent Dropout

Reinforcement Learning with an Abrupt Model Change