Seeing Is Believing: Brain-Inspired Modular Training for Mechanistic Interpretability
We introduce Brain-Inspired Modular Training (BIMT), a method for making neural networks more modular and interpretable. Inspired by brains, BIMT embeds neurons in a geometric space and augments the loss function with a cost proportional to the length of each neuron connection. This is inspired by t...
Na minha lista:
| Principais autores: | , , |
|---|---|
| Formato: | Artigo |
| Idioma: | Inglês |
| Publicado em: |
MDPI AG
2023-12-01
|
| coleção: | Entropy |
| Assuntos: | |
| Acesso em linha: | https://www.mdpi.com/1099-4300/26/1/41 |
| Tags: |
Sem tags, seja o primeiro a adicionar uma tag!
|