Continual Learning with Columnar Spiking Neural Networks

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Larionov, Denis, Bazenkov, Nikolay, Kiselev, Mikhail
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916901810077696
author Larionov, Denis
Bazenkov, Nikolay
Kiselev, Mikhail
author_facet Larionov, Denis
Bazenkov, Nikolay
Kiselev, Mikhail
contents Continual learning is a key feature of biological neural systems, but artificial neural networks often suffer from catastrophic forgetting. Instead of backpropagation, biologically plausible learning algorithms may enable stable continual learning. This study proposes columnar-organized spiking neural networks (SNNs) with local learning rules for continual learning and catastrophic forgetting. Using CoLaNET (Columnar Layered Network), we show that its microcolumns adapt most efficiently to new tasks when they lack shared structure with prior learning. We demonstrate how CoLaNET hyperparameters govern the trade-off between retaining old knowledge (stability) and acquiring new information (plasticity). We evaluate CoLaNET on two benchmarks: Permuted MNIST (ten sequential pixel-permuted tasks) and a two-task MNIST/EMNIST setup. Our model learns ten sequential tasks effectively, maintaining 92% accuracy on each. It shows low forgetting, with only 4% performance degradation on the first task after training on nine subsequent tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2506_17169
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Continual Learning with Columnar Spiking Neural Networks
Larionov, Denis
Bazenkov, Nikolay
Kiselev, Mikhail
Neural and Evolutionary Computing
Artificial Intelligence
Continual learning is a key feature of biological neural systems, but artificial neural networks often suffer from catastrophic forgetting. Instead of backpropagation, biologically plausible learning algorithms may enable stable continual learning. This study proposes columnar-organized spiking neural networks (SNNs) with local learning rules for continual learning and catastrophic forgetting. Using CoLaNET (Columnar Layered Network), we show that its microcolumns adapt most efficiently to new tasks when they lack shared structure with prior learning. We demonstrate how CoLaNET hyperparameters govern the trade-off between retaining old knowledge (stability) and acquiring new information (plasticity). We evaluate CoLaNET on two benchmarks: Permuted MNIST (ten sequential pixel-permuted tasks) and a two-task MNIST/EMNIST setup. Our model learns ten sequential tasks effectively, maintaining 92% accuracy on each. It shows low forgetting, with only 4% performance degradation on the first task after training on nine subsequent tasks.
title Continual Learning with Columnar Spiking Neural Networks
topic Neural and Evolutionary Computing
Artificial Intelligence
url https://arxiv.org/abs/2506.17169