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.

MemControl: Mitigating Memorization in Diffusion Models via Automated Parameter Selection

Capacity Control is an Effective Memorization Mitigation Mechanism in Text-Conditional Diffusion Models

Detecting, Explaining, and Mitigating Memorization in Diffusion Models

Towards Memorization-Free Diffusion Models

Iterative Ensemble Training with Anti-Gradient Control for Mitigating Memorization in Diffusion Models

On Memorization in Diffusion Models

Unveiling and Mitigating Memorization in Text-to-image Diffusion Models through Cross Attention

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

Investigating Data Memorization in 3D Latent Diffusion Models for Medical Image Synthesis

MemBench: Memorized Image Trigger Prompt Dataset for Diffusion Models

Memorized Images in Diffusion Models share a Subspace that can be Located and Deleted

Unconditional Latent Diffusion Models Memorize Patient Imaging Data: Implications for Openly Sharing Synthetic Data

Generative Modeling with Explicit Memory

An Inversion-based Measure of Memorization for Diffusion Models

Finding NeMo: Localizing Neurons Responsible For Memorization in Diffusion Models

Towards a Theoretical Understanding of Memorization in Diffusion Models

Mitigating Memorization In Language Models

Classifier-Free Guidance inside the Attraction Basin May Cause Memorization

Investigating Memorization in Video Diffusion Models

Understanding and Mitigating Memorization in Diffusion Models for Tabular Data