Learning of Non-Monotonic Target Functions by Ising Perceptrons –Learning Curve, Perfect Learning and Perfect Anti-Learning–

DOI10.1143/JPSJ.72.1635MaRDI QIDQ4810554zbMATH OpenFDO

Authors Kanako Tooriyama, Tatsuya Uezu

Publication date 16 August 2004

Published in Journal of the Physical Society of Japan (Search for Journal in Brave)

zbMATH Keywords

replica methodIsing perceptronperfect anti learning

Mathematics Subject Classification ID

Neural networks for/in biological studies, artificial life and related topics (92B20) Time-dependent statistical mechanics (dynamic and nonequilibrium) (82C99)

Recommendations

• Perfect learning and power law in learning from stochastic examples by Ising perceptrons: analysis under one-step replica symmetry breaking ansatz
• On-line learning in the Ising perceptron.
• scientific article; zbMATH DE number 1172290
• A statistical physics approach to learning curves for the inverse Ising problem
• Theoretical analysis of learning dynamics near the opposite singularities in multilayer perceptrons
• Associative learning using Ising-like model
• scientific article; zbMATH DE number 7357537

Cited in

(4)

• Iso-array Acceptors and Learning
• Non-monotonic convergence of online learning algorithms for perceptrons with noisy teacher
• Anti-pairing in learning of a neural network with redundant hidden units
• On-line learning in the Ising perceptron.

This page was built for publication: Learning of Non-Monotonic Target Functions by Ising Perceptrons –Learning Curve, Perfect Learning and Perfect Anti-Learning–

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4810554)