Regularizing with Pseudo-Negatives for Continual Self-Supervised Learning
This work addresses catastrophic forgetting for continual learning systems, representing an incremental improvement with a novel regularization technique.
The paper tackles the problem of catastrophic forgetting in continual self-supervised learning by introducing a Pseudo-Negative Regularization framework, which achieves state-of-the-art performance by balancing plasticity and stability in representation learning.
We introduce a novel Pseudo-Negative Regularization (PNR) framework for effective continual self-supervised learning (CSSL). Our PNR leverages pseudo-negatives obtained through model-based augmentation in a way that newly learned representations may not contradict what has been learned in the past. Specifically, for the InfoNCE-based contrastive learning methods, we define symmetric pseudo-negatives obtained from current and previous models and use them in both main and regularization loss terms. Furthermore, we extend this idea to non-contrastive learning methods which do not inherently rely on negatives. For these methods, a pseudo-negative is defined as the output from the previous model for a differently augmented version of the anchor sample and is asymmetrically applied to the regularization term. Extensive experimental results demonstrate that our PNR framework achieves state-of-the-art performance in representation learning during CSSL by effectively balancing the trade-off between plasticity and stability.