Abstract:Enabling a neural network to sequentially learn multiple tasks is of great significance for expanding the applicability of neural networks in real-world applications. However, artificial neural networks face the well-known problem of catastrophic forgetting. What is worse, the degradation of previously learned skills becomes more severe as the task sequence increases, known as the long-term catastrophic forgetting. It is due to two facts: first, as the model learns more tasks, the intersection of the low-error parameter subspace satisfying for these tasks becomes smaller or even does not exist; second, when the model learns a new task, the cumulative error keeps increasing as the model tries to protect the parameter configuration of previous tasks from interference. Inspired by the memory consolidation mechanism in mammalian brains with synaptic plasticity, we propose a confrontation mechanism in which Adversarial Neural Pruning and synaptic Consolidation (ANPyC) is used to overcome the long-term catastrophic forgetting issue. The neural pruning acts as long-term depression to prune task-irrelevant parameters, while the novel synaptic consolidation acts as long-term potentiation to strengthen task-relevant parameters. During the training, this confrontation achieves a balance in that only crucial parameters remain, and non-significant parameters are freed to learn subsequent tasks. ANPyC avoids forgetting important information and makes the model efficient to learn a large number of tasks. Specifically, the neural pruning iteratively relaxes the current task's parameter conditions to expand the common parameter subspace of the task; the synaptic consolidation strategy, which consists of a structure-aware parameter-importance measurement and an element-wise parameter updating strategy, decreases the cumulative error when learning new tasks. Our approach encourages the synapse to be sparse and polarized, which enables long-term learning and memory. ANPyC exhibits effectiveness and generalization on both image classification and generation tasks with multiple layer perceptron, convolutional neural networks, and generative adversarial networks, and variational autoencoder. The full source code is available at https://github.com/GeoX-Lab/ANPyC.

An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networks

Overcoming Long-Term Catastrophic Forgetting Through Adversarial Neural Pruning and Synaptic Consolidation

Measuring Catastrophic Forgetting in Neural Networks

Does the Adam Optimizer Exacerbate Catastrophic Forgetting?

Catastrophic Forgetting in the Context of Model Updates

Elephant Neural Networks: Born to Be a Continual Learner

Anatomy of Catastrophic Forgetting: Hidden Representations and Task Semantics

Catastrophic Forgetting in Deep Learning: A Comprehensive Taxonomy

Localizing Catastrophic Forgetting in Neural Networks

Overcoming catastrophic forgetting in neural networks

An Improved Dual-Channel Network to Eliminate Catastrophic Forgetting

Maslow's Hammer for Catastrophic Forgetting: Node Re-Use vs Node Activation

Catastrophic forgetting: still a problem for DNNs

Negotiated Representations to Prevent Forgetting in Machine Learning Applications

A Methodology-Oriented Study of Catastrophic Forgetting in Incremental Deep Neural Networks

Understanding Forgetting in Continual Learning with Linear Regression

The Joint Effect of Task Similarity and Overparameterization on Catastrophic Forgetting -- An Analytical Model

Challenging Common Assumptions about Catastrophic Forgetting

Alleviating catastrophic forgetting using context-dependent gating and synaptic stabilization

Catastrophic Importance of Catastrophic Forgetting

Understanding Catastrophic Forgetting and Remembering in Continual Learning with Optimal Relevance Mapping