A building block for hardware belief networks

TL;DR

This paper introduces a transistor-like device as a new hardware building block for probabilistic belief networks, demonstrated through simulated proof-of-concept examples.

Contribution

It proposes a novel hardware component for belief networks, bridging the gap between probabilistic inference and physical hardware implementation.

Findings

Proof-of-concept reciprocal belief network implemented

Proof-of-concept non-reciprocal belief network demonstrated

Device simulation based on experimentally benchmarked models

Abstract

Belief networks represent a powerful approach to problems involving probabilistic inference, but much of the work in this area is software based utilizing standard deterministic hardware based on the transistor which provides the gain and directionality needed to interconnect billions of them into useful networks. This paper proposes a transistor like device that could provide an analogous building block for probabilistic networks. We present two proof-of-concept examples of belief networks, one reciprocal and one non-reciprocal, implemented using the proposed device which is simulated using experimentally benchmarked models.