Biologically inspired sleep algorithm for artificial neural networks
This work addresses incremental learning and generalization issues in ANNs, offering a novel approach inspired by biological sleep, though it is incremental as it builds on existing spiking neural network methods.
The paper tackled catastrophic forgetting and poor generalization in artificial neural networks by implementing a biologically inspired sleep algorithm, demonstrating improved performance on incremental learning tasks, forward transfer, and noise robustness across datasets like MNIST and CUB200.
Sleep plays an important role in incremental learning and consolidation of memories in biological systems. Motivated by the processes that are known to be involved in sleep generation in biological networks, we developed an algorithm that implements a sleep-like phase in artificial neural networks (ANNs). After initial training phase, we convert the ANN to a spiking neural network (SNN) and simulate an offline sleep-like phase using spike-timing dependent plasticity rules to modify synaptic weights. The SNN is then converted back to the ANN and evaluated or trained on new inputs. We demonstrate several performance improvements after applying this processing to ANNs trained on MNIST, CUB200 and a motivating toy dataset. First, in an incremental learning framework, sleep is able to recover older tasks that were otherwise forgotten in the ANN without sleep phase due to catastrophic forgetting. Second, sleep results in forward transfer learning of unseen tasks. Finally, sleep improves generalization ability of the ANNs to classify images with various types of noise. We provide a theoretical basis for the beneficial role of the brain-inspired sleep-like phase for the ANNs and present an algorithmic way for future implementations of the various features of sleep in deep learning ANNs. Overall, these results suggest that biological sleep can help mitigate a number of problems ANNs suffer from, such as poor generalization and catastrophic forgetting for incremental learning.