HAT-CL: A Hard-Attention-to-the-Task PyTorch Library for Continual Learning
This work addresses practical implementation issues for researchers and practitioners in continual learning, though it is incremental as it builds on the existing HAT mechanism.
The authors tackled the challenge of catastrophic forgetting in continual learning by introducing HAT-CL, a user-friendly PyTorch library that redesigns the Hard-Attention-to-the-Task mechanism, achieving improved usability and compatibility with existing networks.
Catastrophic forgetting, the phenomenon in which a neural network loses previously obtained knowledge during the learning of new tasks, poses a significant challenge in continual learning. The Hard-Attention-to-the-Task (HAT) mechanism has shown potential in mitigating this problem, but its practical implementation has been complicated by issues of usability and compatibility, and a lack of support for existing network reuse. In this paper, we introduce HAT-CL, a user-friendly, PyTorch-compatible redesign of the HAT mechanism. HAT-CL not only automates gradient manipulation but also streamlines the transformation of PyTorch modules into HAT modules. It achieves this by providing a comprehensive suite of modules that can be seamlessly integrated into existing architectures. Additionally, HAT-CL offers ready-to-use HAT networks that are smoothly integrated with the TIMM library. Beyond the redesign and reimplementation of HAT, we also introduce novel mask manipulation techniques for HAT, which have consistently shown improvements across various experiments. Our work paves the way for a broader application of the HAT mechanism, opening up new possibilities in continual learning across diverse models and applications.