Volume 96, Number 5, December 2011
|Number of page(s)||6|
|Section||Interdisciplinary Physics and Related Areas of Science and Technology|
|Published online||16 November 2011|
Combined local search strategy for learning in networks of binary synapses
Key Laboratory of Frontiers in Theoretical Physics, Institute of Theoretical Physics, Chinese Academy of Sciences Beijing 100190, China
2 Kavli Institute for Theoretical Physics China, Institute of Theoretical Physics, Chinese Academy of Sciences Beijing 100190, China
3 Department of Physics, The Hong Kong University of Science and Technology - Hong Kong, China
Accepted: 10 October 2011
Learning in networks of binary synapses is known to be an NP-complete problem. A combined stochastic local search strategy in the synaptic weight space is constructed to further improve the learning performance of a single random walker. We apply two correlated random walkers guided by their Hamming distance and associated energy costs (the number of unlearned patterns) to learn a same large set of patterns. Each walker first learns a small part of the whole pattern set (partially different for both walkers but with the same amount of patterns) and then both walkers explore their respective weight spaces cooperatively to find a solution to classify the whole pattern set correctly. The desired solutions locate at the common parts of weight spaces explored by these two walkers. The efficiency of this combined strategy is supported by our extensive numerical simulations and the typical Hamming distance as well as energy cost is estimated by an annealed computation.
PACS: 84.35.+i – Neural networks / 05.40.Fb – Random walks and Levy flights / 75.10.Nr – Spin-glass and other random models
© EPLA, 2011
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.