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.

Reducing catastrophic forgetting of incremental learning in the absence of rehearsal memory with task-specific token

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

An Efficient Strategy for Catastrophic Forgetting Reduction in Incremental Learning

Learning with Recoverable Forgetting

Memory Efficient Data-Free Distillation for Continual Learning.

P2DT: Mitigating Forgetting in task-incremental Learning with progressive prompt Decision Transformer

DKT: Diverse Knowledge Transfer Transformer for Class Incremental Learning

Remembering Transformer for Continual Learning

Model Behavior Preserving for Class-Incremental Learning

Catastrophic Forgetting in the Context of Model Updates

Incremental Embedding Learning with Disentangled Representation Translation.

Few-Shot Self Reminder to Overcome Catastrophic Forgetting

Measuring Catastrophic Forgetting in Neural Networks

Incremental Learning Via Rate Reduction

Prototype Representation Expansion in Incremental Learning

Reducing Catastrophic Forgetting in Online Class Incremental Learning Using Self-Distillation

Class-Incremental Learning via Knowledge Amalgamation

Incremental Learning for Online Data Using QR Factorization on Convolutional Neural Networks

SynapNet: A Complementary Learning System Inspired Algorithm With Real-Time Application in Multimodal Perception

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

Overcoming catastrophic forgetting in neural networks