Bio-inspired computing with memristive devices

21 Sep
Applied Physics Colloquia
Joshua Yang, Department of Electrical and Computer Engineering, University of Massachusetts, Amherst
Friday, September 21, 2018 -
4:00pm to 5:15pm
Pierce 209

Memristive devices1 have become a promising candidate for energy-efficient and high-throughput unconventional computing2. The computing can be implemented on a Resistive Neural Network3 with memristive synapses4 and neurons5 or a Capacitive Neural Network6 with memcapacitive synapses and neurons. In this seminar, I will first briefly introduce memristive devices and the key idea of bio-inspired computing. I will then focus on our recent experimental demonstrations of unconventional computing using memristive networks with different levels of bio-inspirations: first, deep learning accelerators3 with supervised online learning7; second, neuromorphic computing4,8 for pattern classification with unsupervised learning5,6, last, other computing applications such as true random number generators9 for cybersecurity and artificial nociceptors for robotics10.  

1                Yang, J. J. et al. Memristive switching mechanism for metal/oxide/metal nanodevices. Nature Nanotechnology 3, 429-433 (2008).

2                Yang, J. J., Strukov, D. B. & Stewart, D. R. Memristive devices for computing. Nature Nanotechnology 8, 13-24 (2013).

3                Li, C. et al. Analogue signal and image processing with large memristor crossbars. Nature Electronics 1, 52 (2017).

4                Wang, Z. et al. Memristors with diffusive dynamics as synaptic emulators for neuromorphic computing. Nature Materials 16, 101-108 (2017).

5                Wang, Z. et al. Fully memristive neural networks for pattern classification with unsupervised learning. Nature Electronics 1, 137-145 (2018).

6                Wang, Z. et al. Capacitive neural network with neuro-transistors. Nature Communications 9, 3208 (2018).

7                Li, C. et al. Efficient and self-adaptive in-situ learning in multilayer memristive neural networks. Nature communications 9, 2385 (2018).

8                Midya, R. et al. Anatomy of Ag/Hafnia-Based Selectors with 1E10 Nonlinearity. Advanced Materials 29, 1604457 (2017).

9                Jiang, H. et al. A Novel True Random Number Generator Based on a Stochastic Diffusive Memristor. Nature communications 8, 882 (2017).

10              Yoon, J. H. et al. An artificial nociceptor based on a diffusive memristor. Nature Communications 9, 417 (2018).

Speaker Bio: 

Dr. J. Joshua Yang is a professor of the Department of Electrical and Computer Engineering at the University of Massachusetts, Amherst. Before joining UMass in 2015, he spent eight years at HP Labs leading the memristive materials and devices team. His current research interests are Nanoelectronics and Nanoionics for computing applications, where he authored and co-authored over 130 technical papers and holds 97 granted and over 60 pending US Patents. He obtained his PhD from the University of Wisconsin – Madison in Material Science Program in 2007.

Host: 
Donhee Ham
Contact: 
Mike Donohoe