A D Bruce and D Saad 1994 J. Phys. A: Math. Gen. 27 3355 doi:10.1088/0305-4470/27/10/010
A D Bruce and D Saad
Show affiliationsFollowing ideas of Gull, Skilling and MacKay (1992), we develop and explore a statistical-mechanics framework through which one may assign values to the parameters of a model for a 'rule' (instanced, here, by the noisy linear perceptron), on the basis of data instancing the rule. The 'evidence' which the data offers in support of a given assignment, is likened to the free energy of a system with quenched variables (the data): the most probable (MAP) assignments of parameters are those which minimize this free-energy; tracking the free-energy minimum may lead to 'phase transitions' in the preferred assignments. We explore the extent to which the MAP assignments lead to optimal performance.
05.70.Ce Thermodynamic functions and equations of state
07.05.Mh Neural networks, fuzzy logic, artificial intelligence
82C32 Neural nets (See also 68T05, 91E40, 92B20)
82C26 Dynamic and nonequilibrium phase transitions (general)
Issue 10 (21 May 1994)
A D Bruce and D Saad 1994 J. Phys. A: Math. Gen. 27 3355
B Brendebach et al 2009 J. Phys.: Conf. Ser. 190 012186
Adrián A Budini and M O Cáceres 1999 J. Phys. A: Math. Gen. 32 4005
atom
V I Korobov and D Bakalov 2001 J. Phys. B: At. Mol. Opt. Phys. 34 L519
V Nelea et al 2009 J. Phys. D: Appl. Phys. 42 225208
H Hayashi et al 2009 J. Phys.: Conf. Ser. 190 012050
N Zenine et al 2005 J. Phys. A: Math. Gen. 38 9439
M. Tavani 1998 ApJ 497 L21
J Spitaler et al 2009 New J. Phys. 11 113009
M A Gosálvez et al 2009 J. Micromech. Microeng. 19 125011