Neurocomputing and Neurooptimization with Stochastic Memristive Networks

Summary form only given, as follows. The complete presentation was not made available for publication as part of the conference proceedings. Stochastic neural networks have become the state-of-the-art approach for solving machine learning problems, information theory, and statistics. The critical operation in such networks is a stochastic dotproduct. While there have been many demonstrations of dot-product circuits and, separately, stochastic neurons, the efficient hardware implementation combining both functionalities is still missing. In my talk, I will discuss our recent work on addressing this need. I will first discuss compact, fast, energy-efficient, and scalable stochastic dot-product circuits based on passively integrated metal-oxide memristors. The high performance of such circuits is due to mixed-signal implementation. Simultaneously, the efficient stochastic operation is achieved by utilizing the circuit’s noise, intrinsic and/or extrinsic to the memory cell array. The dynamic scaling of weights, enabled by analog memory devices, allows for efficient realization of different annealing approaches to improve functionality. I will then discuss several application demonstrations based on passively integrated TiO2-x memristors, including Hopfield networks for solving combinatorial optimization problems and a Boltzmann machine.