AILGApr 9, 2025

Neuron-level Balance between Stability and Plasticity in Deep Reinforcement Learning

arXiv:2504.08000v1h-index: 34
Originality Incremental advance
AI Analysis

This addresses the challenge of continuous learning in DRL agents, offering a fine-grained approach to balance skill retention and adaptation, though it is incremental as it builds on existing network-level methods.

The paper tackles the stability-plasticity dilemma in deep reinforcement learning by proposing a neuron-level method (NBSP) that identifies skill neurons and applies gradient masking and experience replay to preserve existing skills while adapting to new tasks, showing significant performance improvements on Meta-World and Atari benchmarks.

In contrast to the human ability to continuously acquire knowledge, agents struggle with the stability-plasticity dilemma in deep reinforcement learning (DRL), which refers to the trade-off between retaining existing skills (stability) and learning new knowledge (plasticity). Current methods focus on balancing these two aspects at the network level, lacking sufficient differentiation and fine-grained control of individual neurons. To overcome this limitation, we propose Neuron-level Balance between Stability and Plasticity (NBSP) method, by taking inspiration from the observation that specific neurons are strongly relevant to task-relevant skills. Specifically, NBSP first (1) defines and identifies RL skill neurons that are crucial for knowledge retention through a goal-oriented method, and then (2) introduces a framework by employing gradient masking and experience replay techniques targeting these neurons to preserve the encoded existing skills while enabling adaptation to new tasks. Numerous experimental results on the Meta-World and Atari benchmarks demonstrate that NBSP significantly outperforms existing approaches in balancing stability and plasticity.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes