On the biological plausibility of artificial metaplasticity learning algorithm

The training algorithm studied in this paper is inspired by the biological metaplasticity property of neurons. During the training phase, the Artificial Metaplasticity Learning Algorithm could be considered a new probabilistic version of the presynaptic rule, as during this phase the algorithm assigns higher values for updating the weights in the less probable activations than in the ones with higher probability. The algorithm is proposed for Artificial Neural Networks in general, although results at the moment have only been implemented and tested for Multilayer Perceptrons. Tested on different multidisciplinary applications, experiments show a much more efficient training, improving also Multilayer Perceptron results till the performance of the best systems in the state of the art, systems that usually are much more complex.

[1]  Diego Andina,et al.  Do biological synapses perform probabilistic computations? , 2013, Neurocomputing.

[2]  D. A. Baxter,et al.  In vitro analog of classical conditioning of feeding behavior in aplysia. , 2003, Learning & memory.

[3]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[4]  M. Bear,et al.  Metaplasticity: the plasticity of synaptic plasticity , 1996, Trends in Neurosciences.

[5]  W. Abraham.  Metaplasticity: tuning synapses and networks for plasticity , 2008, Nature Reviews Neuroscience.

[6]  M. Arbib,et al.  A model of cerebellar metaplasticity. , 1998, Learning & memory.

[7]  Wickliffe C Abraham,et al.  NMDA receptor‐mediated metaplasticity during the induction of long‐term depression by low‐frequency stimulation , 2002, The European journal of neuroscience.

[8]  T. Sejnowski,et al.  Storing covariance with nonlinearly interacting neurons , 1977, Journal of mathematical biology.

[9]  Diego Andina,et al.  Artificial Metaplasticity Neural Network Applied to Credit Scoring , 2011, Int. J. Neural Syst..

[10]  Luca Berdondini,et al.  Network Dynamics and Synchronous Activity in cultured Cortical Neurons , 2007, Int. J. Neural Syst..

[11]  D. Debanne,et al.  Long-term plasticity of intrinsic excitability: learning rules and mechanisms. , 2003, Learning & memory.

[12]  Javier Ropero Peláez,et al.  Biological Clues for Up-to-Date Artificial Neurons , 2007 .

[13]  Sang Joon Kim,et al.  A Mathematical Theory of Communication , 2006 .

[14]  Emilio Del-Moral-Hernandez,et al.  A Preliminary Neural Model for Movement Direction Recognition Based on Biologically Plausible Plasticity Rules , 2007, IWINAC.

[15]  Diego Andina,et al.  Probabilistic versus Incremental Presynaptic Learning in Biologically Plausible Synapses , 2011, IWINAC.

[16]  Robert H. Cudmore,et al.  Long-term potentiation of intrinsic excitability in LV visual cortical neurons. , 2004, Journal of neurophysiology.

[17]  Rob Heylen,et al.  Adaptive Thresholds for Neural Networks with Synaptic Noise , 2007, Int. J. Neural Syst..

[18]  W. Singer,et al.  Long-term depression of excitatory synaptic transmission and its relationship to long-term potentiation , 1993, Trends in Neurosciences.

[19]  Joel Quintanilla-Domínguez,et al.  Breast cancer classification applying artificial metaplasticity algorithm , 2011, Neurocomputing.

[20]  Guilherme Neves,et al.  Synaptic plasticity, memory and the hippocampus: a neural network approach to causality , 2008, Nature Reviews Neuroscience.

[21]  Charles F Stevens,et al.  Synaptic plasticity , 1998, Current Biology.

[22]  Shun-ichi Amari,et al.  Mathematical foundations of neurocomputing , 1990, Proc. IEEE.

[23]  D. Andina,et al.  Using neural networks to simulate the Alzheimer’s Disease , 2008, 2008 World Automation Congress.

[24]  E. Bienenstock,et al.  Theory for the development of neuron selectivity: orientation specificity and binocular interaction in visual cortex , 1982, The Journal of neuroscience : the official journal of the Society for Neuroscience.

[25]  T. Bliss,et al.  Synaptic plasticity, memory and the hippocampus: a neural network approach to causality , 2012, Nature Reviews Neuroscience.

[26]  T. Sejnowski,et al.  Homosynaptic long-term depression in hippocampus and mescsrtex , 1990 .

[27]  Diego Andina,et al.  Artificial Metaplasticity can Improve Artificial Neural Networks Learning , 2013, Intell. Autom. Soft Comput..

[28]  D. Andina,et al.  Wood defects classification using Artificial Metaplasticity neural network , 2009, 2009 35th Annual Conference of IEEE Industrial Electronics.