Overcoming Long-term Catastrophic Forgetting through Adversarial Neural Pruning and Synaptic Consolidation

Abstract

Artificial neural networks face the well-known problem of catastrophic forgetting. What's 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. The full source code is available at https://github.com/GeoX-Lab/ANPyC.

14 pages, 11 figures

References in corpus (19)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Auto-Encoding Variational Bayes
• Distilling the Knowledge in a Neural Network
• Conditional Generative Adversarial Nets
• Overcoming catastrophic forgetting in neural networks
• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1
• Progressive Neural Networks
• Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence
• Continual Learning Through Synaptic Intelligence
• PathNet: Evolution Channels Gradient Descent in Super Neural Networks
• Efficient Lifelong Learning with A-GEM
• Progress & Compress: A scalable framework for continual learning
• Alleviating catastrophic forgetting using context-dependent gating and synaptic stabilization
• Overcoming Catastrophic Forgetting by Incremental Moment Matching
• Online Structured Laplace Approximations For Overcoming Catastrophic Forgetting
• Quaternion Gradient and Hessian
• Deep Generative Dual Memory Network for Continual Learning
• Revisiting Natural Gradient for Deep Networks

Cited by in corpus (6)

• Structured Pruning for Deep Convolutional Neural Networks: A survey
• CATRO: Channel Pruning via Class-Aware Trace Ratio Optimization
• Continuous learning of spiking networks trained with local rules
• Sparse-to-dense Feature Matching: Intra and Inter domain Cross-modal Learning in Domain Adaptation for 3D Semantic Segmentation
• Learning by Active Forgetting for Neural Networks
• Tuned Compositional Feature Replays for Efficient Stream Learning