Polymeric Memristors as Entropy Sources for Probabilistic Bit Generation

Abstract

Probabilistic bits, or p-bits, represent a novel computational element that bridges the gap between deterministic classical bits and quantum bits (qubits) used in quantum computing. Unlike classical bits that maintain a definite state of 0 or 1, a p-bit fluctuates between these states in a controlled, stochastic manner. This probabilistic behavior allows for the representation and processing of information in a form that leverages inherent randomness. In this study, a unique approach is presented to generating p-bits using a hybrid conjugated polymer, poly-4-((6-(4H-dithieno[3,2-b:2',3'-d]pyrrol-4-yl)hexyl)oxy)-N,N-diphenylaniline (pTPADTP), as a memristive material. The polymer's conjugated backbone, combined with pendant triphenylamine groups, enables the creation of p-bits through random resistance switching. The stochasticity of this polymeric memristor makes it particularly suited for p-bit applications in stochastic optimization, probabilistic algorithms, and artificial neural networks. The charge transport in the polymer is facilitated by two synergistic percolation mechanisms: one occurring along the polymer backbone and the other through the pendant triphenylamine groups. The study of p-bits generated from pTPADTP opens new avenues for advancing both the theory and practice of computation, where uncertainty and randomness are harnessed as valuable computational resources.