Using Stochastic Spiking Neural Networks on SpiNNaker to Solve Constraint Satisfaction Problems

被引:44
作者
Guerra, Gabriel A. Fonseca [1 ]
Furber, Steve B. [1 ]
机构
[1] Univ Manchester, Sch Comp Sci, Adv Processor Technol Grp, Manchester, Lancs, England
基金
欧洲研究理事会; 英国工程与自然科学研究理事会;
关键词
SpiNNaker; constraint satisfaction; spiking neural networks; stochastic search; spiking neurons; SEARCH; COMPLEXITY; SYSTEM;
D O I
10.3389/fnins.2017.00714
中图分类号
Q189 [神经科学];
学科分类号
071006 ;
摘要
Constraint satisfaction problems (CSP) are at the core of numerous scientific and technological applications. However, CSPs belong to the NP-complete complexity class, for which the existence (or not) of efficient algorithms remains a major unsolved question in computational complexity theory. In the face of this fundamental difficulty heuristics and approximation methods are used to approach instances of NP (e.g., decision and hard optimization problems). The human brain efficiently handles CSPs both in perception and behavior using spiking neural networks (SNNs), and recent studies have demonstrated that the noise embedded within an SNN can be used as a computational resource to solve CSPs. Here, we provide a software framework for the implementation of such noisy neural solvers on the SpiNNaker massively parallel neuromorphic hardware, further demonstrating their potential to implement a stochastic search that solves instances of P and NP problems expressed as CSPs. This facilitates the exploration of new optimization strategies and the understanding of the computational abilities of SNNs. We demonstrate the basic principles of the framework by solving difficult instances of the Sudoku puzzle and of the map color problem, and explore its application to spin glasses. The solver works as a stochastic dynamical system, which is attracted by the configuration that solves the CSP. The noise allows an optimal exploration of the space of configurations, looking for the satisfiability of all the constraints; if applied discontinuously, it can also force the system to leap to a new random configuration effectively causing a restart.
引用
收藏
页数:13
相关论文
共 62 条
[1]  
Alan C., 1965, LOGIC METHODOLOGY PH, P24
[2]   Classic Nintendo games are (computationally) hard [J].
Aloupis, Greg ;
Demaine, Erik D. ;
Guo, Alan ;
Viglietta, Giovanni .
THEORETICAL COMPUTER SCIENCE, 2015, 586 :135-160
[3]  
[Anonymous], 1971, P 3 ANN ACM S THEOR, DOI [DOI 10.1145/800157.805047, 10.1145/800157.805047]
[4]  
[Anonymous], 1999, METAHEURISTICS
[5]  
Appel K. I., 1989, Every planar map is four colorable, V98
[6]   ON THE COMPUTATIONAL-COMPLEXITY OF ISING SPIN-GLASS MODELS [J].
BARAHONA, F .
JOURNAL OF PHYSICS A-MATHEMATICAL AND GENERAL, 1982, 15 (10) :3241-3253
[7]   Neurogrid: A Mixed-Analog-Digital Multichip System for Large-Scale Neural Simulations [J].
Benjamin, Ben Varkey ;
Gao, Peiran ;
McQuinn, Emmett ;
Choudhary, Swadesh ;
Chandrasekaran, Anand R. ;
Bussat, Jean-Marie ;
Alvarez-Icaza, Rodrigo ;
Arthur, John V. ;
Merolla, Paul A. ;
Boahen, Kwabena .
PROCEEDINGS OF THE IEEE, 2014, 102 (05) :699-716
[8]   A graph coloring heuristic using partial solutions and a reactive tabu scheme [J].
Bloechliger, Ivo ;
Zufferey, Nicolas .
COMPUTERS & OPERATIONS RESEARCH, 2008, 35 (03) :960-975
[9]  
Chaitin G. J., 1982, Proceedings of the SIGPLAN Symposium on Compiler Construction, P98, DOI [DOI 10.1145/800230.806984, 10.1145/800230.806984]
[10]   SOME EXPERIMENTS WITH SIMULATED ANNEALING FOR COLORING GRAPHS [J].
CHAMS, M ;
HERTZ, A ;
DEWERRA, D .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 1987, 32 (02) :260-266