Minimum Description Length Hopfield Networks

Published: 01 Jan 2023, Last Modified: 11 May 2025CoRR 2023EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Associative memory architectures are designed for memorization but also offer, through their retrieval method, a form of generalization to unseen inputs: stored memories can be seen as prototypes from this point of view. Focusing on Modern Hopfield Networks (MHN), we show that a large memorization capacity undermines the generalization opportunity. We offer a solution to better optimize this tradeoff. It relies on Minimum Description Length (MDL) to determine during training which memories to store, as well as how many of them.
Loading