1 paper · 1 filter
Aditya Cowsik, Kfir Dolev, Alex Infanger
We present a complete mechanistic description of the algorithm learned by a minimal non-linear sparse data autoencoder in the limit of large input dimension. The model, originally…