Sequential Decision Making Policy

Sequential decision-making policies aim to optimize a sequence of actions over time, maximizing cumulative reward or minimizing risk. Current research focuses on improving the efficiency and scalability of these policies, particularly in high-dimensional action spaces, through techniques like sequential policy factorization and parallel computation methods such as Picard iteration. Significant efforts are also dedicated to enhancing the robustness and generalization of these policies, addressing challenges like offline evaluation, fairness constraints, and the need for reliable confidence estimates in policy comparisons. These advancements have implications for various fields, including supply chain optimization, robotics, and healthcare, where safe and effective decision-making under uncertainty is crucial.

Papers

October 8, 2024

Coevolving with the Other You: Fine-Tuning LLM with Sequential Cooperative Multi-Agent Reinforcement Learning
Hao Ma, Tianyi Hu, Zhiqiang Pu, Boyin Liu, Xiaolin Ai, Yanyan Liang, Min Chen
Large Language Model Reinforcement Learning Fine Tuning Multi Agent Reinforcement Learning LLM Fine Tuning Adaptive Coevolution Sequential Decision Making Policy

July 8, 2024

CANDID DAC: Leveraging Coupled Action Dimensions with Importance Differences in DAC
Philipp Bordne, M. Asif Hasan, Eddie Bergman, Noor Awad, André Biedenkapp
Action Space Importance Score Algorithm Configuration Sequential Decision Making Policy

June 4, 2024

Speeding up Policy Simulation in Supply Chain RL
Vivek Farias, Joren Gijsbrechts, Aryan Khojandi, Tianyi Peng, Andrew Zheng
Policy OpTimization Supply Chain Based Policy Sequential Decision Making Policy Backward Chaining

May 27, 2024

OPERA: Automatic Offline Policy Evaluation with Re-weighted Aggregates of Multiple Estimators
Allen Nie, Yash Chandak, Christina J. Yuan, Anirudhan Badrinath, Yannis Flet-Berliac, Emma Brunskil
Policy Evaluation Linear Estimator Weight Aggregation Offline Policy Evaluation Sequential Decision Making Policy

August 29, 2023

Improving Generalization in Reinforcement Learning Training Regimes for Social Robot Navigation
Adam Sigal, Hsiu-Chin Lin, AJung Moon
Reinforcement Learning Strong Generalization Social Navigation Autonomous Mobile Robot Social Robot Navigation Sequential Decision Making Policy

December 30, 2022

Risk-Sensitive Policy with Distributional Reinforcement Learning
Thibaut Théate, Damien Ernst
Reinforcement Learning Distributional Reinforcement Learning Sequential Decision Making Policy Risk Aware Policy

December 8, 2022

Model-based trajectory stitching for improved behavioural cloning and its applications
Charles A. Hepburn, Giovanni Montana
Financial Application Offline Reinforcement Learning Behavior Cloning Offline Planning Behavior Policy Sequential Decision Making Policy Trajectory Corner Case

November 8, 2022

Reinforcement Learning with Stepwise Fairness Constraints
Zhun Deng, He Sun, Zhiwei Steven Wu, Linjun Zhang, David C. Parkes
Reinforcement Learning Fairness Constraint Tabular Reinforcement Learning Sequential Decision Making Policy

October 25, 2022

In-context Reinforcement Learning with Algorithm Distillation
Michael Laskin, Luyu Wang, Junhyuk Oh, Emilio Parisotto, Stephen Spencer, Richie Steigerwald, DJ Strouse, Steven Hansen, Angelos Filos, Ethan Brooks, Maxime Gazeau, Himanshu Sahni, Satinder Singh, Volodymyr Mnih
Reinforcement Learning Data Efficient Sequential Decision Making Policy Algorithm Distillation

May 22, 2022

Offline Policy Comparison with Confidence: Benchmarks and Baselines
Anurag Koul, Mariano Phielipp, Alan Fern
New Benchmark Offline Reinforcement Learning Baseline Result High Confidence Comparative Evaluation Baseline Model Offline Policy Evaluation Sequential Decision Making Policy

May 11, 2022

Developing cooperative policies for multi-stage reinforcement learning tasks
Jordan Erskine, Chris Lehnert
Reinforcement Learning Hierarchical Reinforcement Learning Long Horizon Task Maze Environment Sequential Decision Making Policy

March 29, 2022

Best Arm Identification in Restless Markov Multi-Armed Bandits
P. N. Karthik, Kota Srinivas Reddy, Vincent Y. F. Tan
Multi Armed Bandit Markov Chain Best Arm Identification Sequential Decision Making Policy