ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2307.03186
22
12

TGRL: An Algorithm for Teacher Guided Reinforcement Learning

6 July 2023
Idan Shenfeld
Zhang-Wei Hong
Aviv Tamar
Pulkit Agrawal
ArXivPDFHTML
Abstract

Learning from rewards (i.e., reinforcement learning or RL) and learning to imitate a teacher (i.e., teacher-student learning) are two established approaches for solving sequential decision-making problems. To combine the benefits of these different forms of learning, it is common to train a policy to maximize a combination of reinforcement and teacher-student learning objectives. However, without a principled method to balance these objectives, prior work used heuristics and problem-specific hyperparameter searches to balance the two objectives. We present a principled\textit{principled}principled approach, along with an approximate implementation for dynamically\textit{dynamically}dynamically and automatically\textit{automatically}automatically balancing when to follow the teacher and when to use rewards. The main idea is to adjust the importance of teacher supervision by comparing the agent's performance to the counterfactual scenario of the agent learning without teacher supervision and only from rewards. If using teacher supervision improves performance, the importance of teacher supervision is increased and otherwise it is decreased. Our method, Teacher Guided Reinforcement Learning\textit{Teacher Guided Reinforcement Learning}Teacher Guided Reinforcement Learning (TGRL), outperforms strong baselines across diverse domains without hyper-parameter tuning.

View on arXiv
Comments on this paper