Do Not Forget to Attend to Uncertainty while Mitigating Catastrophic Forgetting

Abstract

One of the major limitations of deep learning models is that they face catastrophic forgetting in an incremental learning scenario. There have been several approaches proposed to tackle the problem of incremental learning. Most of these methods are based on knowledge distillation and do not adequately utilize the information provided by older task models, such as uncertainty estimation in predictions. The predictive uncertainty provides the distributional information can be applied to mitigate catastrophic forgetting in a deep learning framework. In the proposed work, we consider a Bayesian formulation to obtain the data and model uncertainties. We also incorporate self-attention framework to address the incremental learning problem. We define distillation losses in terms of aleatoric uncertainty and self-attention. In the proposed work, we investigate different ablation analyses on these losses. Furthermore, we are able to obtain better results in terms of accuracy on standard benchmarks.

arXiv commentAccepted WACV 2021

References in corpus (10)

• Distilling the Knowledge in a Neural Network
• Overcoming catastrophic forgetting in neural networks
• Self-Attention Generative Adversarial Networks
• Continual Learning Through Synaptic Intelligence
• Alleviating catastrophic forgetting using context-dependent gating and synaptic stabilization
• Overcoming Catastrophic Forgetting by Incremental Moment Matching
• Generative replay with feedback connections as a general strategy for continual learning
• Incremental Classifier Learning with Generative Adversarial Networks
• SupportNet: solving catastrophic forgetting in class incremental learning with support data
• Bayesian Incremental Learning for Deep Neural Networks