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.

AFEC: Active Forgetting of Negative Transfer in Continual Learning

Progressive Learning without Forgetting

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

Learning After Learning: Positive Backward Transfer in Continual Learning

Learning by Active Forgetting for Neural Networks

Learning with Recoverable Forgetting

Adaptive Rentention & Correction for Continual Learning

Anti-retroactive Interference for Lifelong Learning

Bio-inspired, task-free continual learning through activity regularization

Continual Recognition with Adaptive Memory Update.

Lifelong Learning With Cycle Memory Networks

Incorporating neuro-inspired adaptability for continual learning in artificial intelligence

Reinforced Continual Learning

Accurate Forgetting for Heterogeneous Federated Continual Learning

Overcome Anterograde Forgetting with Cycled Memory Networks

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

Continual learning under domain transfer with sparse synaptic bursting

Beyond Not-Forgetting: Continual Learning with Backward Knowledge Transfer

Adaptive Plasticity Improvement for Continual Learning

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

Class Incremental Learning with Less Forgetting Direction and Equilibrium Point