Self-activating Neural Ensembles For Continual 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, and thus depend on human supervision. Our task-agnostic method, Self-Activating Neural Ensembles (SANE), uses a modular architecture designed to avoid catastrophic forgetting without making any such assumptions. At the beginning of each trajectory, a module in the SANE ensemble is activated to determine the agent's next policy. During training, new modules are created as needed and only activated modules are updated to ensure that unused modules remain unchanged. This system enables our method to retain and leverage old skills, while growing and learning new ones. We demonstrate our approach on visually rich procedurally generated environments.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keypowers2022self

Related papers

• Modular Continual Learning In A Unified Visual Environment (2017)0.00
• Meta-reinforcement Learning With Self-modifying Networks (2022)4.52
• Continual Reinforcement Learning With Complex Synapses (2018)0.00
• Continual Learning Using World Models For Pseudo-rehearsal (2019)0.00
• Lifelong Reinforcement Learning Via Neuromodulation (2024)0.00
• Learning Without Time-based Embodiment Resets In Soft-actor Critic (2025)0.00
• Automatic Goal Generation For Reinforcement Learning Agents (2017)0.00
• Learning Synthetic Environments For Reinforcement Learning With Evolution Strategies (2021)0.00