Statistical mechanics of learning: A variational approach for real data

Physical Review Letters

Volumn 89, Issue 10, 2002, Pages 1083021-1083024

  Malzahn, Dörthe ^a   Opper, Manfred ^a  

^a ASTON UNIVERSITY   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

ARTIFICIAL NEURAL NETWORK; PROCEDURES; STATISTICS; THEORETICAL MODEL;

MODELS, THEORETICAL; NEURAL NETWORKS (COMPUTER); STATISTICS AS TOPIC;

APPROXIMATION THEORY; DATA REDUCTION; ERROR ANALYSIS; HAMILTONIANS; LEARNING ALGORITHMS; LEARNING SYSTEMS; MARKOV PROCESSES; PROBABILITY DISTRIBUTIONS; SET THEORY; VARIATIONAL TECHNIQUES;

DATA SETS;

STATISTICAL MECHANICS;

EID: 19044377144     PISSN: 00319007     EISSN: None     Source Type: Journal    
DOI: 10.1103/PhysRevLett.89.108302     Document Type: Article

References (15)

1. A. Engel and C. Van den Broeck, Statistical Mechanics of Learning (Cambridge University, Cambridge, 2001).
2. H. Nishimori, Statistical Physics of Spin Glasses and Information Processing (Oxford Science Publications, Oxford, 2001).
3. M. Mézard, G. Parisi, and M. A. Virasoro, Spin Glass Theory and Beyond, Lecture Notes in Physics Vol. 9 (World Scientific, Singapore, 1987).
4. G. Wahba, Splines Models for Observational Data, Series in Applied Mathematics Vol 59 (SIAM, Philadelphia, 1990).
5. R. Neal, Bayesian Learning for Neural Networks, Lecture Notes in Statistics Vol. 118 (Springer, New York, 1996).
6. C. K. I. Williams and C. E. Rasmussen, in NIPS, edited by D. S. Touretzky, M. C. Mozer, and M. E. Hasselmo (MIT, Cambridge, MA, 1996), Vol. 8, p. 514.
7. W. Bialek, C. G. Callan, and S. P. Strong, Phys. Rev. Lett. 77, 4693 (1996).
8. See, for example, Advances in Kernel Methods Support Vector Learning, edited by B. Schölkopf, C. J. C. Burges, and A. J. Smola (MIT, Cambridge, MA, 1999).
9. M. Opper and O. Winther, Neural Comput. 12, 2655 (2000).
10. M. Mézard and G. Parisi, J. Phys. I (France) 1, 809 (1991).
11. R. P. Feynman and A. R. Hibbs, Quantum Mechanics and Path Integrals (McGraw-Hill, New York, 1965).
12. Similarly, one can compute sample fluctuations.
13. The data sets can be downloaded from http://www1.ics.uci.edu/mlearn/MLSummary.html.
14. E. Gardner, J. Phys. A 21, 257 (1988).
15. One can model decision surfaces by polynomials of arbitrary order.