A spiking recurrent neural network with phase change memory synapses for decision making

被引：2

作者：

Pedretti, G. ^{[1
,2
]}

Milo, V ^{[1
,2
]}

Hashemkhani, S. ^{[1
,2
]}

Mannocci, P. ^{[1
,2
]}

Melnic, O. ^{[1
,2
]}

Chicca, E. ^{[3
,4
]}

Ielmini, D. ^{[1
,2
]}

机构：

[1] Politecn Milan, Dipartimento Elettron Informaz & Bioingn, I-20133 Milan, Italy

[2] IU NET, I-20133 Milan, Italy

[3] Bielefeld Univ, Fac Technol, Bielefeld, Germany

[4] Bielefeld Univ, Cognit Interact Technol Ctr Excellence CITEC, Bielefeld, Germany

来源：

2020 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS) | 2020年

基金：

欧洲研究理事会;

关键词：

phase change memory (PCM); Hebbian learning; spiking recurrent neural network (RNN); associative memory; decision making; constraint satisfaction problems (CSPs); NEURONS;

D O I：

10.1109/iscas45731.2020.9180513

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Neuronal activity of recurrent neural networks (RNNs) experimentally observed in the hippocampus is widely believed to play a key role for mammalian ability to associate concepts and make decisions. For this reason, RNNs have rapidly gained strong interest as computational enabler of brain-inspired cognitive functions in hardware. From the technology viewpoint, nonvolatile memory devices such as phase change memory (PCM) and resistive switching memory (RRAM) have become a key asset to allow for high synaptic density and biorealistic cognitive functionality. In this work, we demonstrate for the first time associative learning and decision making in a hardware Hopfield RNN with 6 spiking neurons and PCM synapses via storage, recall and competition of attractor states. We also experimentally demonstrate the solution of a constraint satisfaction problem (CSP) namely a Sudoku with size 2x2 in hardware and 9x9 in simulation. These results support spiking RNNs with PCM devices for the implementation of decision making capabilities in hardware neuromorphic systems.

引用

页数：5

共 24 条

[1]

Amit DJ., 1989, Modeling Brain Function: The World of Attractor Neural Networks

[2]

Binas J, 2016, IEEE INT SYMP CIRC S, P2094, DOI 10.1109/ISCAS.2016.7538992

[3]

Cai F., 2019, arXiv preprint arXiv:1903.11194

[4] A VLSI recurrent network of integrate-and-fire neurons connected by plastic synapses with long-term memory [J].

Chicca, E ;

Badoni, D ;

Dante, V ;

D'Andreagiovanni, M ;

Salina, G ;

Carota, L ;

Fusi, S ;

Del Giudice, P .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 2003, 14 (05) :1297-1307

[5]

Corradi F, 2015, IEEE INT SYMP CIRC S, P2708, DOI 10.1109/ISCAS.2015.7169245

[6] A memristive plasticity model of voltage-based STDP suitable for recurrent bidirectional neural networks in the hippocampus [J].

Diederich, Nick ;

Bartsch, Thorsten ;

Kohlstedt, Hermann ;

Ziegler, Martin .

SCIENTIFIC REPORTS, 2018, 8

[7] Brain-like associative learning using a nanoscale non-volatile phase change synaptic device array [J].

Eryilmaz, Sukru B. ;

Kuzum, Duygu ;

Jeyasingh, Rakesh ;

Kim, SangBum ;

BrightSky, Matthew ;

Lam, Chung ;

Wong, H. -S. Philip .

FRONTIERS IN NEUROSCIENCE, 2014, 8

[8] Real time unsupervised learning of visual stimuli in neuromorphic VLSI systems [J].

Giulioni, Massimiliano ;

Corradi, Federico ;

Dante, Vittorio ;

del Giudice, Paolo .

SCIENTIFIC REPORTS, 2015, 5

[9] Searching for memories, Sudoku, implicit check bits, and the iterative use of not-always-correct rapid neural computation [J].

Hopfield, J. J. .

NEURAL COMPUTATION, 2008, 20 (05) :1119-1164

[10] NEURAL NETWORKS AND PHYSICAL SYSTEMS WITH EMERGENT COLLECTIVE COMPUTATIONAL ABILITIES [J].

HOPFIELD, JJ .

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA-BIOLOGICAL SCIENCES, 1982, 79 (08) :2554-2558

← 1 2 3 →