Recognition Capabilities of a Hopfield Model with Auxiliary Hidden Neurons

TL;DR

This paper investigates an enhanced Hopfield model with auxiliary hidden neurons, demonstrating improved recognition capacity and stability over the traditional model, especially at high memory loads.

Contribution

It introduces a Hopfield model with hidden layers derived via Hubbard-Stratonovich transformation, showing increased storage capacity and robustness.

Findings

Recognition capacity is significantly increased.

The model avoids abrupt failure at high memory loads.

It lacks a naturally defined basin of attraction.

Abstract

We study the recognition capabilities of the Hopfield model with auxiliary hidden layers, which emerge naturally upon a Hubbard-Stratonovich transformation. We show that the recognition capabilities of such a model at zero-temperature outperform those of the original Hopfield model, due to a substantial increase of the storage capacity and the lack of a naturally defined basin of attraction. The modified model does not fall abruptly in a regime of complete confusion when memory load exceeds a sharp threshold.