The Alternative Computing Group at NIST has an ongoing program in developing prototype circuits and designing new architectures for neuromorphic computing utilizing exotic devices such as metal-oxide and phase-change memristors, magnetic tunnel junctions, and Josephson junctions. These devices are combined with custom-designed conventional CMOS circuits to realize diverse functionalities, from crossbar-based machine learning to race-logic-based DNA sequencing.  Opportunities exist for experimental work in device fabrication and measurement using NIST’s state-of-the-art NanoFab and extensive measurement capabilities, CMOS circuit design work for foundry tape-out, and theoretical work developing new algorithms and architectures that leverage the low-energy, high-speed properties of emerging devices.  The ultimate goal is to explore the landscape of emerging hardware-based artificial intelligence systems to better understand the role played by measurement and metrology in these complex, adaptive systems.     

"Streaming Batch Eigenupdates for Hardware Neuromorphic Networks," B. D. Hoskins, M. W. Daniels, S. Huang, A. Madhavan, G. C. Adam, N. Zhitenev, J. J. McClelland, and M. D. Stiles, arXiv:1903.01635 [cs.LG]

"Neuromorphic computing with nanoscale spintronic oscillators," J. Torrejon, M. Riou, F. A. Araujo, S. Tsunegi, G. Khalsa, D. Querlioz, P. Bortolotti, V. Cros, K. Yakushiji, A. Fukushima, H. Kubota, S. Yuasa, M. D. Stiles, and J. Grollier, Nature 547, 428–431 (2017).

"Stateful characterization of resistive switching TiO­2 with electron beam induced currents," B. D. Hoskins, G. C. Adam, E. Strelcov, N. Zhitenev, A. Kolmakov, D. B. Strukov, and J. J. McClelland, Nature Communications 8, 1972 (2017).

key words

Neuromorphic computing; artificial intelligence; memristors; RRAM; CMOS; machine learning; cognitive computing; phase change memory; superconducting circuits; magnetic tunnel junctions; race logic

Citizenship:  Open to U.S. citizens

Level:  Open to Postdoctoral applicants