Learning in Wilson-Cowan model for metapopulation
This work addresses the challenge of using neural mass models for machine learning, though it appears incremental as it combines the model with existing architectures like CNNs and transformers.
The paper tackled the problem of adapting the Wilson-Cowan model for metapopulation into a learning algorithm for classification tasks, achieving high accuracy on datasets like MNIST, CIFAR-10, and IMDB.
The Wilson-Cowan model for metapopulation, a Neural Mass Network Model, treats different subcortical regions of the brain as connected nodes, with connections representing various types of structural, functional, or effective neuronal connectivity between these regions. Each region comprises interacting populations of excitatory and inhibitory cells, consistent with the standard Wilson-Cowan model. By incorporating stable attractors into such a metapopulation model's dynamics, we transform it into a learning algorithm capable of achieving high image and text classification accuracy. We test it on MNIST and Fashion MNIST, in combination with convolutional neural networks, on CIFAR-10 and TF-FLOWERS, and, in combination with a transformer architecture (BERT), on IMDB, always showing high classification accuracy. These numerical evaluations illustrate that minimal modifications to the Wilson-Cowan model for metapopulation can reveal unique and previously unobserved dynamics.