Uncertainty-guided Continual Learning with Bayesian Neural Networks

Abstract

Continual learning aims to learn new tasks without forgetting previously learned ones. This is especially challenging when one cannot access data from previous tasks and when the model has a fixed capacity. Current regularization-based continual learning algorithms need an external representation and extra computation to measure the parameters' \textit{importance}. In contrast, we propose Uncertainty-guided Continual Bayesian Neural Networks (UCB), where the learning rate adapts according to the uncertainty defined in the probability distribution of the weights in networks. Uncertainty is a natural way to identify \textit{what to remember} and \textit{what to change} as we continually learn, and thus mitigate catastrophic forgetting. We also show a variant of our model, which uses uncertainty for weight pruning and retains task performance after pruning by saving binary masks per tasks. We evaluate our UCB approach extensively on diverse object classification datasets with short and long sequences of tasks and report superior or on-par performance compared to existing approaches. Additionally, we show that our model does not necessarily need task information at test time, i.e. it does not presume knowledge of which task a sample belongs to.

arXiv commentAccepted at ICLR 2020

References in corpus (10)

• Overcoming catastrophic forgetting in neural networks
• Pruning Convolutional Neural Networks for Resource Efficient Inference
• Continual Learning Through Synaptic Intelligence
• PathNet: Evolution Channels Gradient Descent in Super Neural Networks
• Efficient Lifelong Learning with A-GEM
• Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networks
• Overcoming Catastrophic Forgetting by Incremental Moment Matching
• Less-forgetting Learning in Deep Neural Networks
• Uncertainty Estimations by Softplus normalization in Bayesian Convolutional Neural Networks with Variational Inference
• Fast and Scalable Bayesian Deep Learning by Weight-Perturbation in Adam

Cited by in corpus (5)

• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• On Sequential Bayesian Inference for Continual Learning
• A Simple yet Effective Framework for Active Learning to Rank
• Continual Learning via Bit-Level Information Preserving
• Compression-aware Continual Learning using Singular Value Decomposition