Similarity-based context aware continual learning for spiking neural networks

TL;DR

A Similarity-based Context Aware Spiking Neural Network (SCA-SNN) continual learning algorithm to efficiently accomplish task incremental learning and class incremental learning and adaptively select similar groups of neurons for related tasks is proposed.

Abstract

Biological brains have the capability to adaptively coordinate relevant neuronal populations based on the task context to learn continuously changing tasks in real-world environments. However, existing spiking neural network-based continual learning algorithms treat each task equally, ignoring the guiding role of different task similarity associations for network learning, which limits knowledge utilization efficiency. Inspired by the context-dependent plasticity mechanism of the brain, we propose a Similarity-based Context Aware Spiking Neural Network (SCA-SNN) continual learning algorithm to efficiently accomplish task incremental learning and class incremental learning. Based on contextual similarity across tasks, the SCA-SNN model can adaptively reuse neurons from previous tasks that are beneficial for new tasks (the more similar, the more neurons are reused) and flexibly expand new neurons for the new task (the more similar, the fewer neurons are expanded). Selective reuse and discriminative expansion significantly improve the utilization of previous knowledge and reduce energy consumption. Extensive experimental results on CIFAR100, ImageNet generalized datasets, and FMNIST-MNIST, SVHN-CIFAR100 mixed datasets show that our SCA-SNN model achieves superior performance compared to both SNN-based and DNN-based continual learning algorithms. Additionally, our algorithm has the capability to adaptively select similar groups of neurons for related tasks, offering a promising approach to enhancing the biological interpretability of efficient continual learning.

Figures (4)

• Figure 1: The architecture of SCA-SNN. The SCA-SNN first evaluates the similarity between the new task and each old task, then discriminatively extends new neurons and selectively reuses learned neurons based on the degree of new task similarity.
• Figure 2: The performance comparison of SCA-SNN on SNN-based continual learning in task incremental learning.
• Figure 3: The effect on the performance of different similarity assessments (a) and the parameters of proposed similarity (b), neuronal discriminative expansion (c) and selective reuse (d).
• Figure 4: Similarity-based neuronal population assignment. a): Colored blocks represent activated convolutional feature maps, and white blocks represent pruned feature maps. The fork blocks are placeholders only. b): The relationship between task-to-task distance and the pruning rate of new task pathways on neurons belonging to old tasks.