Tunable Synaptic Working Memory with Volatile Memristive Devices

TL;DR

This paper demonstrates how volatile memristive devices can be electrically tuned to implement flexible, reconfigurable working memory in neuromorphic hardware, capable of matching various timescales for cognitive tasks.

Contribution

It introduces a novel approach using silver-based memristive devices with tunable parameters to realize adaptable working memory in hardware and simulations.

Findings

Memristive devices' retention time and switching probability are electrically tunable.

Successful demonstration of working memory in hardware for associative tasks.

Simulations show applicability in biological and symbolic contexts.

Abstract

Different real-world cognitive tasks evolve on different relevant timescales. Processing these tasks requires memory mechanisms able to match their specific time constants. In particular, the working memory utilizes mechanisms that span orders of magnitudes of timescales, from milliseconds to seconds or even minutes. This plentitude of timescales is an essential ingredient of working memory tasks like visual or language processing. This degree of flexibility is challenging in analog computing hardware because it requires the integration of several reconfigurable capacitors of different size. Emerging volatile memristive devices present a compact and appealing solution to reproduce reconfigurable temporal dynamics in a neuromorphic network. We present a demonstration of working memory using a silver-based memristive device whose key parameters, retention time and switching probability, can be electrically tuned and adapted to the task at hand. First, we demonstrate the principles of working memory in a small scale hardware to execute an associative memory task. Then, we use the experimental data in two larger scale simulations, the first featuring working memory in a biological environment, the second demonstrating associative symbolic working memory.