Pseudo-Recursal: Solving the Catastrophic Forgetting Problem in Deep Neural Networks

Abstract

In general, neural networks are not currently capable of learning tasks in a sequential fashion. When a novel, unrelated task is learnt by a neural network, it substantially forgets how to solve previously learnt tasks. One of the original solutions to this problem is pseudo-rehearsal, which involves learning the new task while rehearsing generated items representative of the previous task/s. This is very effective for simple tasks. However, pseudo-rehearsal has not yet been successfully applied to very complex tasks because in these tasks it is difficult to generate representative items. We accomplish pseudo-rehearsal by using a Generative Adversarial Network to generate items so that our deep network can learn to sequentially classify the CIFAR-10, SVHN and MNIST datasets. After training on all tasks, our network loses only 1.67% absolute accuracy on CIFAR-10 and gains 0.24% absolute accuracy on SVHN. Our model's performance is a substantial improvement compared to the current state of the art solution.

References in corpus (3)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• FearNet: Brain-Inspired Model for Incremental Learning
• Measuring Catastrophic Forgetting in Neural Networks

Cited by in corpus (16)

• A continual learning survey: Defying forgetting in classification tasks
• Generative replay with feedback connections as a general strategy for continual learning
• Pseudo-Rehearsal: Achieving Deep Reinforcement Learning without Catastrophic Forgetting
• Efficient Continual Learning in Neural Networks with Embedding Regularization
• Replay in Deep Learning: Current Approaches and Missing Biological Elements
• Continuous learning of spiking networks trained with local rules
• AirLoop: Lifelong Loop Closure Detection
• Dynamic Sub-graph Distillation for Robust Semi-supervised Continual Learning
• Analyzing the Forgetting Problem in the Pretrain-Finetuning of Dialogue Response Models
• Variational Prototype Replays for Continual Learning
• Automatic Recall Machines: Internal Replay, Continual Learning and the Brain
• A Conceptual Framework for Lifelong Learning
• Domain-aware Neural Language Models for Speech Recognition
• Tuned Compositional Feature Replays for Efficient Stream Learning
• GRIm-RePR: Prioritising Generating Important Features for Pseudo-Rehearsal
• ZS-IL: Looking Back on Learned Experiences For Zero-Shot Incremental Learning