Do Neural Networks Lose Plasticity in a Gradually Changing World?
This addresses the issue of neural networks failing to adapt over time in real-world scenarios for continual learning applications, but it is incremental as it builds on existing plasticity research.
The paper tackles the problem of neural networks losing plasticity in continual learning by investigating gradually changing environments instead of abrupt task transitions, showing that loss of plasticity is largely mitigated in such settings.
Continual learning has become a trending topic in machine learning. Recent studies have discovered an interesting phenomenon called loss of plasticity, referring to neural networks gradually losing the ability to learn new tasks. However, existing plasticity research largely relies on contrived settings with abrupt task transitions, which often do not reflect real-world environments. In this paper, we propose to investigate a gradually changing environment, and we simulate this by input/output interpolation and task sampling. We perform theoretical and empirical analysis, showing that the loss of plasticity is an artifact of abrupt tasks changes in the environment and can be largely mitigated if the world changes gradually.