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.

Anti-retroactive Interference for Lifelong Learning

Progressive Learning without Forgetting

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

Learning with Recoverable Forgetting

Memory Aware Synapses: Learning what (not) to forget

Dual Memory Model for Experience-Once Task-Incremental Lifelong Learning.

Overcome Anterograde Forgetting with Cycled Memory Networks

Lifelong Learning With Cycle Memory Networks

Adaptive Rentention & Correction for Continual Learning

Schematic Memory Persistence and Transience for Efficient and Robust Continual Learning

Incorporating neuro-inspired adaptability for continual learning in artificial intelligence

AdaER: An Adaptive Experience Replay Approach for Continual Lifelong Learning

Memory Recall: A Simple Neural Network Training Framework Against Catastrophic Forgetting

AFEC: Active Forgetting of Negative Transfer in Continual Learning

Drift to Remember

Task Agnostic Continual Learning via Meta Learning

A Simple Lifelong Learning Approach

Learn to Grow: A Continual Structure Learning Framework for Overcoming Catastrophic Forgetting

Online Continual Learning Via the Meta-learning Update with Multi-scale Knowledge Distillation and Data Augmentation

Adaptive online continual multi-view learning

Lifelong Neural Predictive Coding: Learning Cumulatively Online without Forgetting