Recurrent networks of coupled Winner-Take-All oscillators for solving constraint satisfaction problems

We present a recurrent neuronal network, modeled as a continuous-time dynamical system, that can solve constraint satisfaction problems. Discrete variables are represented by coupled Winner-Take-All (WTA) networks, and their values are encoded in localized patterns of oscillations that are learned by the recurrent weights in these networks. Constraints over the variables are encoded in the network connectivity. Although there are no sources of noise, the network can escape from local optima in its search for solutions that satisfy all constraints by modifying the effective network connectivity through oscillations. If there is no solution that satisfies all constraints, the network state changes in a seemingly random manner and its trajectory approximates a sampling procedure that selects a variable assignment with a probability that increases with the fraction of constraints satisfied by this assignment. External evidence, or input to the network, can force variables to specific values. When new inputs are applied, the network re-evaluates the entire set of variables in its search for states that satisfy the maximum number of constraints, while being consistent with the external input. Our results demonstrate that the proposed network architecture can perform a deterministic search for the optimal solution to problems with non-convex cost functions. The network is inspired by canonical microcircuit models of the cortex and suggests possible dynamical mechanisms to solve constraint satisfaction problems that can be present in biological networks, or implemented in neuromorphic electronic circuits.

[1]  Kevan A. C. Martin,et al.  A Canonical Microcircuit for Neocortex , 1989, Neural Computation.

[2]  Ueli Rutishauser,et al.  State-Dependent Computation Using Coupled Recurrent Networks , 2008, Neural Computation.

[3]  Ramón Huerta,et al.  Dynamical encoding by networks of competing neuron groups: winnerless competition. , 2001 .

[4]  Maurizio Mattia,et al.  Collective Behavior of Networks with Linear (VLSI) Integrate-and-Fire Neurons , 1999, Neural Computation.

[5]  Xiao-Jing Wang.  Decision Making in Recurrent Neuronal Circuits , 2008, Neuron.

[6]  R. Douglas,et al.  Neuronal circuits of the neocortex. , 2004, Annual review of neuroscience.

[7]  Ramón Huerta,et al.  Transient Cognitive Dynamics, Metastability, and Decision Making , 2008, PLoS Comput. Biol..

[8]  K. Lashley.  The problem of serial order in behavior , 1951 .

[9]  R Huerta,et al.  Dynamical encoding by networks of competing neuron groups: winnerless competition. , 2001, Physical review letters.

[10]  S Dehaene,et al.  Neural networks that learn temporal sequences by selection. , 1987, Proceedings of the National Academy of Sciences of the United States of America.

[11]  V. Zhigulin,et al.  On the origin of reproducible sequential activity in neural circuits. , 2004, Chaos.

[12]  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.

[13]  G. Schöner.  The Cambridge Handbook of Computational Psychology: Dynamical Systems Approaches to Cognition , 2008 .

[14]  Alex M Thomson,et al.  Excitatory and inhibitory connections show selectivity in the neocortex , 2005, The Journal of physiology.

[15]  C. Stevens,et al.  Voltage dependence of NMDA-activated macroscopic conductances predicted by single-channel kinetics , 1990, The Journal of neuroscience : the official journal of the Society for Neuroscience.

[16]  Giacomo Indiveri,et al.  A Current-Mode Hysteretic Winner-take-all Network, with Excitatory and Inhibitory Coupling , 2001 .

[17]  A. Thomson,et al.  Target and temporal pattern selection at neocortical synapses. , 2002, Philosophical transactions of the Royal Society of London. Series B, Biological sciences.

[18]  Jean-Jacques E. Slotine,et al.  Collective Stability of Networks of Winner-Take-All Circuits , 2011, Neural Computation.

[19]  N. Brunel,et al.  Calcium-based plasticity model explains sensitivity of synaptic changes to spike pattern, rate, and dendritic location , 2012, Proceedings of the National Academy of Sciences.

[20]  P. J. Sjöström,et al.  Rate, Timing, and Cooperativity Jointly Determine Cortical Synaptic Plasticity , 2001, Neuron.

[21]  Kazuyuki Aihara,et al.  Chaotic simulated annealing by a neural network model with transient chaos , 1995, Neural Networks.

[22]  J. Tanji.  Sequential organization of multiple movements: involvement of cortical motor areas. , 2001, Annual review of neuroscience.

[23]  Jean-Jacques E. Slotine,et al.  Competition Through Selective Inhibitory Synchrony , 2012, Neural Computation.

[24]  G. Karmos,et al.  Entrainment of Neuronal Oscillations as a Mechanism of Attentional Selection , 2008, Science.

[25]  A Aertsen,et al.  Accurate Spike Synchronization in Cortex , 1998, Zeitschrift fur Naturforschung. C, Journal of biosciences.

[26]  Kevan A. C. Martin,et al.  Whose Cortical Column Would that Be? , 2010, Front. Neuroanat..

[27]  John Lazzaro,et al.  Winner-Take-All Networks of O(N) Complexity , 1988, NIPS.

[28]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[29]  Bard Ermentrout,et al.  A model for complex sequence learning and reproduction in neural populations , 2011, Journal of Computational Neuroscience.

[30]  S. Amari.  Dynamics of pattern formation in lateral-inhibition type neural fields , 1977, Biological Cybernetics.

[31]  Bruce W. Knight,et al.  Dynamics of Encoding in Neuron Populations: Some General Mathematical Features , 2000, Neural Computation.

[32]  D J Amit,et al.  Neural networks counting chimes. , 1988, Proceedings of the National Academy of Sciences of the United States of America.

[33]  Xiao-Jing Wang,et al.  A Biophysically Based Neural Model of Matching Law Behavior: Melioration by Stochastic Synapses , 2006, The Journal of Neuroscience.

[34]  Carver A. Mead,et al.  Neuromorphic electronic systems , 1990, Proc. IEEE.

[35]  Gustavo Deco,et al.  Sequential Memory: A Putative Neural and Synaptic Dynamical Mechanism , 2005, Journal of Cognitive Neuroscience.

[36]  Jerry A. Fodor,et al.  Information and association , 1986, Notre Dame J. Formal Log..

[37]  James P. Crutchfield.  Critical Computation, Phase Transitions, and Hierarchical Learning , 2001 .

[38]  Stefan Habenschuss,et al.  Stochastic Computations in Cortical Microcircuit Models , 2013, PLoS Comput. Biol..

[39]  J. J. Hopfield,et al.  “Neural” computation of decisions in optimization problems , 1985, Biological Cybernetics.

[40]  Kevan A C Martin,et al.  Neuroanatomy: Uninhibited Connectivity in Neocortex? , 2011, Current Biology.

[41]  P. Somogyi,et al.  Glutamate decarboxylase‐immunoreactive terminals of Golgi‐impregnated axoaxonic cells and of presumed basket cells in synaptic contact with pyramidal neurons of the cat's visual cortex , 1983, The Journal of comparative neurology.

[42]  Gerald M. Edelman,et al.  Temporal sequence learning in winner-take-all networks of spiking neurons demonstrated in a brain-based device , 2013, Front. Neurorobot..

[43]  J J Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities. , 1982, Proceedings of the National Academy of Sciences of the United States of America.

[44]  Xiao-Jing Wang,et al.  Probabilistic Decision Making by Slow Reverberation in Cortical Circuits , 2002, Neuron.

[45]  Richard Hans Robert Hahnloser,et al.  An ultra-sparse code underliesthe generation of neural sequences in a songbird , 2002, Nature.

[46]  Ron Sun,et al.  The Cambridge Handbook of Computational Psychology , 2008 .

[47]  W. Singer,et al.  Modulation of Neuronal Interactions Through Neuronal Synchronization , 2007, Science.

[48]  Professor Moshe Abeles,et al.  Local Cortical Circuits , 1982, Studies of Brain Function.

[49]  P. Fries.  A mechanism for cognitive dynamics: neuronal communication through neuronal coherence , 2005, Trends in Cognitive Sciences.

[50]  John J Hopfield,et al.  Neurodynamics of mental exploration , 2009, Proceedings of the National Academy of Sciences.

[51]  R. Douglas,et al.  A Quantitative Map of the Circuit of Cat Primary Visual Cortex , 2004, The Journal of Neuroscience.

[52]  B. Kamgar-Parsi.  Dynamical stability and parameter selection in neural optimization , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.

[53]  Gert Cauwenberghs,et al.  Neuromorphic Silicon Neuron Circuits , 2011, Front. Neurosci.

[54]  Richard Hans Robert Hahnloser,et al.  Spike-Time-Dependent Plasticity and Heterosynaptic Competition Organize Networks to Produce Long Scale-Free Sequences of Neural Activity , 2010, Neuron.

[55]  Marimuthu Palaniswami,et al.  Neural techniques for combinatorial optimization with applications , 1998, IEEE Trans. Neural Networks.

[56]  Bard Ermentrout,et al.  Complex dynamics in winner-take-all neural nets with slow inhibition , 1992, Neural Networks.

[57]  Jeffrey S. Johnson,et al.  The Dynamic Field Theory and Embodied Cognitive Dynamics , 2008 .

[58]  József Fiser,et al.  Spontaneous Cortical Activity Reveals Hallmarks of an Optimal Internal Model of the Environment , 2011, Science.

[59]  M. Abeles.  Local Cortical Circuits: An Electrophysiological Study , 1982 .

[60]  D. Amit,et al.  Model of global spontaneous activity and local structured activity during delay periods in the cerebral cortex. , 1997, Cerebral cortex.

[61]  M. Mattia,et al.  Population dynamics of interacting spiking neurons. , 2002, Physical review. E, Statistical, nonlinear, and soft matter physics.

[62]  H Sompolinsky,et al.  Associative neural network model for the generation of temporal patterns. Theory and application to central pattern generators. , 1988, Biophysical journal.

[63]  D. McCormick,et al.  Comparative electrophysiology of pyramidal and sparsely spiny stellate neurons of the neocortex. , 1985, Journal of neurophysiology.

[64]  Xiao-Jing Wang.  Neurophysiological and computational principles of cortical rhythms in cognition. , 2010, Physiological reviews.

[65]  Yun Wang,et al.  Synaptic connections and small circuits involving excitatory and inhibitory neurons in layers 2-5 of adult rat and cat neocortex: triple intracellular recordings and biocytin labelling in vitro. , 2002, Cerebral cortex.

[66]  Daniel J. Amit,et al.  Modeling brain function: the world of attractor neural networks, 1st Edition , 1989 .

[67]  Gregor Schöner,et al.  An embodied account of serial order: How instabilities drive sequence generation , 2010, Neural Networks.

[68]  Mahesan Niranjan,et al.  A theoretical investigation into the performance of the Hopfield model , 1990, IEEE Trans. Neural Networks.

[69]  R. Douglas,et al.  A functional microcircuit for cat visual cortex. , 1991, The Journal of physiology.

[70]  Peter C. Cheeseman,et al.  Where the Really Hard Problems Are , 1991, IJCAI.

[71]  F. Pulvermüller,et al.  Spatiotemporal Signatures of Large-Scale Synfire Chains for Speech Processing as Revealed by MEG , 2008, Cerebral cortex.

[72]  H. Markram,et al.  Interneurons of the neocortical inhibitory system , 2004, Nature Reviews Neuroscience.