Raffaele Marino, Lorenzo Buffoni, Lorenzo Chicchi, Francesca Di Patti, Diego Febbe, Lorenzo Giambagli, Duccio Fanelli
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.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Sentiment Analysis | IMDb | Accuracy | 87.46 | Bert+ Wilson-Cowan model RNN |
| Image Classification | Fashion-MNIST | Accuracy | 91.35 | CNN+ Wilson-Cowan model RNN |
| Image Classification | Fashion-MNIST | Accuracy | 88.39 | Wilson-Cowan model RNN |
| Image Classification | Flowers (Tensorflow) | Accuracy | 84.85 | CNN+ Wilson-Cowan model RNN |
| Image Classification | CIFAR-10 | Percentage correct | 86.59 | CNN+ Wilson-Cowan model RNN |
| Image Classification | MNIST | Accuracy | 99.31 | CNN+ Wilson-Cowan model RNN |
| Image Classification | MNIST | Accuracy | 98.13 | Wilson-Cowan model RNN |