Self-Activating Neural Ensembles for Continual Reinforcement Learning

Sam Powers, Abhinav Gupta

Keywords: continual reinforcement learning, lifelong learning, deep reinforcement learning

Abstract: The ability for an agent to continuously learn new skills without catastrophically forgetting existing knowledge is of critical importance for the development of generally intelligent agents. Most methods devised to address this problem depend heavily on well-defined task boundaries which simplify the problem considerably. Our task-agnostic method, Self-Activating Neural Ensembles (SANE), uses a hierarchical modular architecture designed to avoid catastrophic forgetting without making any such assumptions. At each timestep a path through the SANE tree is activated; during training only activated nodes are updated, ensuring that unused nodes do not undergo catastrophic forgetting. Additionally, new nodes are created as needed, allowing the system to leverage and retain old skills while growing and learning new ones. We demonstrate our approach on MNIST and a set of grid world environments, demonstrating that SANE does not undergo catastrophic forgetting where existing methods do.

One-sentence Summary: We present a novel tree-structured neural architecture that enables the learning of tasks sequentially.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2301.00141/code)

Reviewed Version (pdf): https://openreview.net/references/pdf?id=XGFy4bTT8