Mutual information based knowledge transfer under state-action dimension mismatch

Michael Wan, Tanmay Gangwani, Jian Peng

Research output: Contribution to conferencePaperpeer-review


Deep reinforcement learning (RL) algorithms have achieved great success on a wide variety of sequential decision-making tasks. However, many of these algorithms suffer from high sample complexity when learning from scratch using environmental rewards, due to issues such as credit-assignment and high-variance gradients, among others. Transfer learning, in which knowledge gained on a source task is applied to more efficiently learn a different but related target task, is a promising approach to improve the sample complexity in RL. Prior work has considered using pre-trained teacher policies to enhance the learning of the student policy, albeit with the constraint that the teacher and the student MDPs share the state-space or the action-space. In this paper, we propose a new framework for transfer learning where the teacher and the student can have arbitrarily different state- and action-spaces. To handle this mismatch, we produce embeddings which can systematically extract knowledge from the teacher policy and value networks, and blend it into the student networks. To train the embeddings, we use a task-aligned loss and show that the representations could be enriched further by adding a mutual information loss. Using a set of challenging simulated robotic locomotion tasks involving many-legged centipedes, we demonstrate successful transfer learning in situations when the teacher and student have different state- and action-spaces.

Original languageEnglish (US)
Number of pages10
StatePublished - 2020
Event36th Conference on Uncertainty in Artificial Intelligence, UAI 2020 - Virtual, Online
Duration: Aug 3 2020Aug 6 2020


Conference36th Conference on Uncertainty in Artificial Intelligence, UAI 2020
CityVirtual, Online

ASJC Scopus subject areas

  • Artificial Intelligence


Dive into the research topics of 'Mutual information based knowledge transfer under state-action dimension mismatch'. Together they form a unique fingerprint.

Cite this