ROLE OF NOISES IN NEURAL NETWORKS

Authors
ALBEVERIO S FENG JF QIAN MP
Citation
S. Albeverio et al., ROLE OF NOISES IN NEURAL NETWORKS, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics, 52(6), 1995, pp. 6593-6606
Citations number
32
Categorie Soggetti
Physycs, Mathematical","Phsycs, Fluid & Plasmas
Journal title
Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics → ACNP
ISSN journal
1063651X
Volume
52
Issue
6
Year of publication
1995
Part
B
Pages
6593 - 6606
Database
ISI
SICI code
1063-651X(1995)52:6<6593:RONINN>2.0.ZU;2-5
Abstract
The important role played by noise in retrieving memories in a network is discussed. Two-stage annealing is proposed to retrieve memories st ored in a neural network. The network, undergoing two-stage annealing, behaves like a nonergodic system, and the noise helps the network to select memory inside a local region in which the initial stimuli drop. Theoretical and numerical results of two-stage annealing are presente d, and some further possible applications are pointed out.