Efficient and self-adaptive in-situ learning in multilayer memristor neural networks
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.
Document Details
- Document Type
- Pub Defense Publication
- Publication Date
- Jun 19, 2018
- Source ID
- 10.1038/s41467-018-04484-2
Entities
People
- Can Li
- Daniel Belkin
- Eric Montgomery
- Hao Jiang
- J. Joshua Yang
- John Paul Strachan
- Mark Barnell
- Miao Hu
- Ning Ge
- Peng Lin
- Peng Yan
- Qiangfei Xia
- Qing Wu
- R. Stanley Williams
- Wenhao Song
- Yunning Li
- Zhongrui Wang
Organizations
- Air Force Research Laboratory
- National Science Foundation