Face classification using electronic synapses

TLDR

Conventional hardware platforms consume large energy due to data movement between processor and off‑chip memory, whereas brain‑inspired analog weight storage enables more efficient cognitive tasks. The authors aim to present an analogue non‑volatile resistive memory (electronic synapse) fabricated with foundry‑friendly materials. The device operates as an analogue non‑volatile resistive memory using these materials, enabling bidirectional continuous weight modulation. The 1024‑cell array achieves near‑CPU accuracy on grey‑scale face classification with per‑iteration energy consumption 1,000× lower, demonstrating the feasibility of energy‑efficient large‑scale neuromorphic systems.

Abstract

Abstract Conventional hardware platforms consume huge amount of energy for cognitive learning due to the data movement between the processor and the off-chip memory. Brain-inspired device technologies using analogue weight storage allow to complete cognitive tasks more efficiently. Here we present an analogue non-volatile resistive memory (an electronic synapse) with foundry friendly materials. The device shows bidirectional continuous weight modulation behaviour. Grey-scale face classification is experimentally demonstrated using an integrated 1024-cell array with parallel online training. The energy consumption within the analogue synapses for each iteration is 1,000 × (20 ×) lower compared to an implementation using Intel Xeon Phi processor with off-chip memory (with hypothetical on-chip digital resistive random access memory). The accuracy on test sets is close to the result using a central processing unit. These experimental results consolidate the feasibility of analogue synaptic array and pave the way toward building an energy efficient and large-scale neuromorphic system.