Efficient and self-adaptive in-situ learning in multilayer memristor neural networks

TLDR

Memristors with tunable resistance states are emerging building blocks of artificial neural networks, yet large‑scale in‑situ learning in multilayer networks remains unproven due to device engineering and circuit integration challenges. The authors monolithically integrate hafnium oxide memristors with a foundry‑made transistor array to form a multilayer neural network. We experimentally demonstrate in‑situ learning and competitive classification accuracy on a standard dataset, and simulations indicate that scaling the network would further improve accuracy, confirming the memristor neural network as a high‑speed, energy‑efficient AI platform.

Abstract

Memristors with tunable resistance states are emerging building blocks of artificial neural networks. However, in situ learning on a large-scale multiple-layer memristor network has yet to be demonstrated because of challenges in device property engineering and circuit integration. Here we monolithically integrate hafnium oxide-based memristors with a foundry-made transistor array into a multiple-layer neural network. We experimentally demonstrate in situ learning capability and achieve competitive classification accuracy on a standard machine learning dataset, which further confirms that the training algorithm allows the network to adapt to hardware imperfections. Our simulation using the experimental parameters suggests that a larger network would further increase the classification accuracy. The memristor neural network is a promising hardware platform for artificial intelligence with high speed-energy efficiency.