Model-Based Episodic Memory Induces Dynamic Hybrid Controls

Hung Le, Thommen Karimpanal George, Majid Abdolshah, Truyen Tran, Svetha Venkatesh

Keywords: Reinforcement learning, episodic memory, sample-efficiency

TL;DR: A model-based episodic control facilitating complementary learning systems

Abstract: Episodic control enables sample efficiency in reinforcement learning by recalling past experiences from an episodic memory. We propose a new model-based episodic memory of trajectories addressing current limitations of episodic control. Our memory estimates trajectory values, guiding the agent towards good policies. Built upon the memory, we construct a complementary learning model via a dynamic hybrid control unifying model-based, episodic and habitual learning into a single architecture. Experiments demonstrate that our model allows significantly faster and better learning than other strong reinforcement learning agents across a variety of environments including stochastic and non-Markovian settings.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

Code: https://github.com/thaihungle/MBEC-plus

Community Implementations: [ 5 code implementations](https://www.catalyzex.com/paper/arxiv:2111.02104/code)