A Novel Stochastic Learning Rule for Neural Networks
The purpose of this article is the introduction of a novel stochastic Hebb-like learning rule for neural networks which combines features of unsupervised (Hebbian) and supervised (reinforcement) learning. This learning rule is stochastic with respect to the selection of the time points when a synaptic modification is induced by simultantious activation of the pre- and postsynaptic neuron. Moreover, the learning rule does not only affect the synapse between pre- and postsynaptic neuron which is called homosynaptic plasticity but effects also further remote synapses of the pre- and postsynaptic neuron. This more complex form of plasticity has recently come into the light of interest of experimental investigations in neurobiology and is called heterosynaptic plasticity. Our learning rule is motivated by these experimental findings and gives a qualitative explanation of this kind of synaptic plasticity. Additionally, we give some numerical results that demonstrate that our learning rule works well in training neural networks, even in the presence of noise.
KeywordsNeural Network Hide Layer Synaptic Weight Postsynaptic Neuron Reinforcement Signal
Unable to display preview. Download preview PDF.
- 1.Bak, P., Chialvo, D.R.: Adaptive Learning by Extremal Dynamics and Negative Feedback. Phys. Rev. E 63, 031912-1–031912-12 (2001)Google Scholar
- 3.Bliss, T.V.P., Lomo, T.: Long-lasting Potentiation of Synaptic Transmission in the Dentate Area of the Anaesthetized Rabbit Following Stimulation of the Perforant Path. J. Physiol. 232, 331–356 (1973)Google Scholar
- 7.Emmert-Streib, F.: Aktive Computation in Offenen Systemen. Lerndynamiken in biologischen Systemen: Vom Netzwerk zum Organismus. Dissertation, Universität Bremen, Mensch & Buch Verlag (2003)Google Scholar
- 8.Emmert-Streib, F.: Active Learning in Recurrent Neural Networks Facilitated by a Hebb-like Learning Rule with Memory. Neural Information Processing - Letters and Reviews 9(2), 31–40 (2005)Google Scholar
- 11.Hebb, D.O.: The Organization of Behavior. Wiley, New York (1949)Google Scholar
- 12.Hertz, J., Krogh, A., Palmer, R.G.: Introduction to the Theory of Neural Computation. Addison-Wesley, Reading (1991)Google Scholar
- 13.Holland, J.: Adaptation in Natural and Artificial Systems. University of Michigan Press, Ann Arbor (1975)Google Scholar
- 16.Koch, C.: Biophysics of Computation. Oxford Press (1999)Google Scholar
- 18.Otmakhova, N.A., Lisman, J.E.: D1/D5 Dopamine Receptors Inhibit Depotentiation of CA1 Synapses via cAMP-dependent Mechanism. J. Neuroscience 18, 1270–1279 (1998)Google Scholar
- 21.Werbos, P.: Beyond Regression: New tools for Prediction and Analysis in the Behavioral Sciences. Ph.D. Thesis, Harvard University (1974)Google Scholar