Maximizing competition in a biologically plausible neural network
The goal of this work is increasing competition among the neurons of a one layer feed-forward neural network. The feed-forward weights from input to the cells are learnt using a Hebbian based learning rule and the weights between the neurons are learnt with an anti-Hebbian rule. A dynamic learning rate is introduced and it is claimed that this rule improves the competition.
Download file
Counter downloads: 377
Keywords
neural networks, Hebbian learning, anti-Hebbian learning, Oja rule, covariance learning, competition, нейронные сети, правило обучения Хэбба, антихэббовское обучение, правило Ойя, ковариантное обучение, конкуренцияAuthors
| Name | Organization | |
| Kermani Arash K. | Technical University of Chemnitz | arash@hrz.tu-chemnitz.de |
| Spitsyn Vladimir G. | Tomsk Polytechnic University | spvg@tpu.ru |
| Hamker Fred H. | Technical University of Chemnitz | fred.hamker@informatik.tu-chemnitz.de |
References
Haxby J.V., Hoffmann E.A., and Gobbini M.I. Human Neural Systems for Face recognition and Social Communication // Society of Biological Psychiatry. 2002. No. 51. P. 59-67.
Dayan P., Abbot L.F. Theoretical Neuroscience. Cambridge. Massachusetts: MIT Press, 2001. P. 45-122.
Wiltschut J., Hamker F. Efficient coding correlates with spatial frequency tuning in a model of V1 receptive filed organization // Visual Neuroscience. 2009. No. 26. P. 21-34.
Hebb D.O. Organization of behavior. New York: Wiley, 1949. 335 p.
Foldiak P. Forming sparse representations by local anti-Hebbian learning // Biological Cybernetics. 1990. No. 64. P. 165-170.
Hubel D.H. and Wiesel T.N. Receptive fields, binocular interaction and functional architecture in the cat's visual cortex // J. Physiology. 1961. No. 160. P. 106-154.
Olshausen B.A. and Field D.J. Emergence of simple-cell receptive field properties by learning a sparse code for natural images // Nature. 1996. No. 381. P. 607-609.
Oja E. Simplified neuron model as a principal component analyzer // J. Mathematical Biology. November 1982. No. 15 (3). P. 267-273.
Sejnowski T.J. Storing covariance with nonlinearity interacting neurons // J. Mathematical Biology. 1997. No. 4. P. 303-321.
Robert C.M., Mark F.B. LTP and LTD: An embarrassment of Riches // Neuron. 2004. No. 44. P. 5-21.
Teichmann M., Wiltschut J., Hamker F.H., Learning invariance from natural images inspired by observations in the primary visual cortex // Neural Computation. 2012. No. 24(5). P. 1271-1296.
Hoyer P.O. Non-negative matrix factorization with sparseness constraints // J. Machine Learning Research. 2004. No. 5. P. 1457-1469.
