Continual Learning Using Bayesian Neural Networks

TL;DR

This paper introduces Continual Bayesian Learning Networks (CBLN), a method using Bayesian Neural Networks to mitigate catastrophic forgetting in continual learning by allocating resources dynamically without needing past data.

Contribution

The paper proposes CBLN, a novel Bayesian neural network approach that maintains task-specific distributions and adapts resource allocation to prevent forgetting.

Findings

Addresses catastrophic forgetting effectively

Does not require access to past training data

Shows promising results on MNIST and UCR datasets

Abstract

Continual learning models allow to learn and adapt to new changes and tasks over time. However, in continual and sequential learning scenarios in which the models are trained using different data with various distributions, neural networks tend to forget the previously learned knowledge. This phenomenon is often referred to as catastrophic forgetting. The catastrophic forgetting is an inevitable problem in continual learning models for dynamic environments. To address this issue, we propose a method, called Continual Bayesian Learning Networks (CBLN), which enables the networks to allocate additional resources to adapt to new tasks without forgetting the previously learned tasks. Using a Bayesian Neural Network, CBLN maintains a mixture of Gaussian posterior distributions that are associated with different tasks. The proposed method tries to optimise the number of resources that are needed to learn each task and avoids an exponential increase in the number of resources that are involved in learning multiple tasks. The proposed method does not need to access the past training data and can choose suitable weights to classify the data points during the test time automatically based on an uncertainty criterion. We have evaluated our method on the MNIST and UCR time-series datasets. The evaluation results show that our method can address the catastrophic forgetting problem at a promising rate compared to the state-of-the-art models.