Reconfigurable Stochastic neurons based on tin oxide/MoS2 hetero-memristors for simulated annealing and the Boltzmann machine

Xiaodong Yan,Matthew Chin,Han Wang,Wei Wu,Mike Shuo-Wei Chen,Zhihan Zhang,Tong Wu,Jiangbin Wu,Jing Guo,Aoyang Zhang,Madan Dubey,Jiahui Ma

doi:10.1038/s41467-021-26012-5

Abstract

Neuromorphic hardware implementation of Boltzmann Machine using a network of stochastic neurons can allow non-deterministic polynomial-time (NP) hard combinatorial optimization problems to be efficiently solved. Efficient implementation of such Boltzmann Machine with simulated annealing desires the statistical parameters of the stochastic neurons to be dynamically tunable, however, there has been limited research on stochastic semiconductor devices with controllable statistical distributions. Here, we demonstrate a reconfigurable tin oxide (SnOx)/molybdenum disulfide (MoS2) heterogeneous memristive device that can realize tunable stochastic dynamics in its output sampling characteristics. The device can sample exponential-class sigmoidal distributions analogous to the Fermi-Dirac distribution of physical systems with quantitatively defined tunable “temperature” effect. A BM composed of these tunable stochastic neuron devices, which can enable simulated annealing with designed “cooling” strategies, is conducted to solve the MAX-SAT, a representative in NP-hard combinatorial optimization problems. Quantitative insights into the effect of different “cooling” strategies on improving the BM optimization process efficiency are also provided.

Highlights

Neuromorphic hardware implementation of Boltzmann Machine using a network of stochastic neurons can allow non-deterministic polynomial-time (NP) hard combinatorial optimization problems to be efficiently solved
Stochastic neuron devices are essential for the neural network implementation of key emerging non-von-Neumann computing concepts such as the Boltzmann machines, which are recurrent artificial neural networks with stochastic features analogous to the thermodynamics of real-world physical systems
We overcome such challenge with a three-terminal stochastic hetero-memristor based on tin oxide/MoS2 heterostructure, which demonstrates tunable statistical distributions enabled by the gate modulation

Summary

Introduction

Neuromorphic hardware implementation of Boltzmann Machine using a network of stochastic neurons can allow non-deterministic polynomial-time (NP) hard combinatorial optimization problems to be efficiently solved. To realize the hardware implementation of the BM that can allow the “temperature” control and the precise execution of desired “cooling” strategy, it is essential to have electronic devices that can generate exponential-class stochastic sampling with dynamically tunable distribution parameters.

Results

Conclusion