Tractable Approximations for Probabilistic Models: The Adaptive Thouless-Anderson-Palmer Mean Field Approach
- 23 April 2001
- journal article
- research article
- Published by American Physical Society (APS) in Physical Review Letters
- Vol. 86 (17) , 3695-3699
- https://doi.org/10.1103/physrevlett.86.3695
Abstract
We develop an advanced mean field method for approximating averages in probabilistic data models that is based on the Thouless-Anderson-Palmer (TAP) approach of disorder physics. In contrast to conventional TAP, where the knowledge of the distribution of couplings between the random variables is required, our method adapts to the concrete couplings. We demonstrate the validity of our approach, which is so far restricted to models with nonglassy behavior, by replica calculations for a wide class of models as well as by simulations for a real data set.Keywords
All Related Versions
This publication has 16 references indexed in Scilit:
- Gaussian Processes for Classification: Mean-Field AlgorithmsNeural Computation, 2000
- Information geometry and Plefka's mean-field theoryJournal of Physics A: General Physics, 2000
- Belief propagation vs. TAP for decoding corrupted messagesEurophysics Letters, 1998
- Mean-field theory of Boltzmann machine learningPhysical Review E, 1998
- Mean Field Approach to Bayes Learning in Feed-Forward Neural NetworksPhysical Review Letters, 1996
- Microscopic Equations and Stability Conditions in Optimal Neural NetworksEurophysics Letters, 1995
- The space of interactions in neural networks: Gardner's computation with the cavity methodJournal of Physics A: General Physics, 1989
- Mean-Field Equations for the Matching and the Travelling Salesman ProblemsEurophysics Letters, 1986
- Solution of 'Solvable model of a spin glass'Philosophical Magazine, 1977
- Solvable Model of a Spin-GlassPhysical Review Letters, 1975