Continual Learning via Low-Rank Network Updates
Continual learning seeks to train a single network for multiple tasks (one after another), where training data for each task is only available during the training of that task. Neural networks tend to forget older tasks when they are trained for the newer tasks; this property is often known as catastrophic forgetting. To address this issue, continual learning methods use episodic memory, parameter regularization, masking and pruning, or extensible network structures. In this paper, we propose a new continual learning framework based on low-rank factorization. In particular, we represent the network weights for each layer as a linear combination of several low-rank (or rank-1) matrices. To update the network for a new task, we learn a low-rank (or rank-1) matrix and add that to the weights of every layer. We also introduce an additional selector vector that assigns different weights to the low-rank matrices learned for the previous tasks. We show that our approach performs better than the current state-of-the-art methods in terms of accuracy and forgetting. Our method also offers better memory efficiency compared to episodic memory-based approaches.
PDF Abstract