State sampling dependence of Hopfield network inference

DOI10.1088/0253-6102/57/1/27MaRDI QIDQ2913543zbMATH OpenOpenAlexFDO

Authors Hai-Ping Huang

Publication date 26 September 2012

Published in Communications in Theoretical Physics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1104.4931

zbMATH Keywords

Hopfield networkinferencespin glass

Mathematics Subject Classification ID

General topics in artificial intelligence (68T01) Neural nets applied to problems in time-dependent statistical mechanics (82C32) Neural networks for/in biological studies, artificial life and related topics (92B20)

Abstract: The fully connected Hopfield network is inferred based on observed magnetizations and pairwise correlations. We present the system in the glassy phase with low temperature and high memory load. We find that the inference error is very sensitive to the form of state sampling. When a single state is sampled to compute magnetizations and correlations, the inference error is almost indistinguishable irrespective of the sampled state. However, the error can be greatly reduced if the data is collected with state transitions. Our result holds for different disorder samples and accounts for the previously observed large fluctuations of inference error at low temperatures.

Recommendations

• Network inference with hidden units
• The appropriateness of ignorance in the inverse kinetic Ising model
• Typical error pattern recovery of the Hopfield memory under error-tolerant conditions
• A binary relation inference network Part 2. State-space techniques
• THE INVESTIGATION OF THE SPIN GLASS PROPERTIES OF THE HOPFIELD NEURAL NETWORK MODEL

This page was built for publication: State sampling dependence of Hopfield network inference

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2913543)