Knowledge Transfer From Teachers To Learners In Growing-batch Reinforcement Learning

Abstract

Standard approaches to sequential decision-making exploit an agent's ability to continually interact with its environment and improve its control policy. However, due to safety, ethical, and practicality constraints, this type of trial-and-error experimentation is often infeasible in many real-world domains such as healthcare and robotics. Instead, control policies in these domains are typically trained offline from previously logged data or in a growing-batch manner. In this setting a fixed policy is deployed to the environment and used to gather an entire batch of new data before being aggregated with past batches and used to update the policy. This improvement cycle can then be repeated multiple times. While a limited number of such cycles is feasible in real-world domains, the quality and diversity of the resulting data are much lower than in the standard continually-interacting approach. However, data collection in these domains is often performed in conjunction with human experts

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyemedomnnamdi2023knowledge

Related papers

• Iterative Batch Reinforcement Learning Via Safe Diversified Model-based Policy Search (2024)0.00
• TGRL: An Algorithm For Teacher Guided Reinforcement Learning (2023)0.00
• Improving Interactive Reinforcement Learning: What Makes A Good Teacher? (2019)11.19
• Post-convergence Sim-to-real Policy Transfer: A Principled Alternative To Cherry-picking (2025)0.00
• Continual Policy Distillation From Distributed Reinforcement Learning Teachers (2026)0.00
• Learning Curricula In Open-ended Worlds (2023)0.00
• Stackelberg Batch Policy Learning (2023)0.00
• Guided Online Distillation: Promoting Safe Reinforcement Learning By Offline Demonstration (2023)4.52