Memristive Devices for Neuromorphic Computing : vTools Events

IEEE.org | IEEE Xplore Digital Library | IEEE Standards | IEEE Spectrum | More Sites

Memristive Devices for Neuromorphic Computing

#Memristive #Devices #Neuromorphic #Computing #Memory

Northwestern ECE Distinguished Seminar Series

Detailed Information: https://www.mccormick.northwestern.edu/electrical-computer/news-events/events/

Professor J. Joshua Yang, University of Southern California

CMOS technology has been the mainstream hardware technology for the development of ubiquitous information technology so far. In the era of ‘big data’ and ‘Internet of Things’ nowadays, the traditional computing architecture based on CMOS hardware has become increasingly inefficient to support Artificial Intelligence (AI) and Machine Learning (ML), which necessitates some emerging technologies, such as memristive technology. Memristive devices(1) have become a promising candidate to enable bio-inspired computing with much improved efficiency and throughput(2,3). Such computing can be implemented on a Resistive Neural Network(4) with memristive synapses(5) and neurons(6) or a Capacitive Neural Network(7,8) with memcapacitive synapses and neurons. I will first briefly introduce the promises and challenges of memristive materials and devices for those applications and then discuss examples with different levels of bio-inspiration: first, deep learning accelerators(9) with supervised online learning(10); second, neuromorphic computing for pattern classification with unsupervised learning(6); last, other computing applications, such as reinforcement learning(11) for decision making, artificial nociceptors for robotics(12), provable key destruction(13) and true random number generators(14) for cybersecurity.

1 Nature Nanotechnology 8, 13 (2013)

2 Nature materials 18, 309 (2019)

3 Nature Reviews Materials 5, 173 (2020)

4 Nature Machine Intelligence 1, 434 (2019)

5 Nature Materials 16, 101 (2017)

6 Nature Electronics 1, 137 (2018)

7 Nature Communications 9, 3208 (2018)

8 Nature Machine Intelligence 1, 49 (2019)

9 Nature Electronics 1, 52 (2018)

10 Nature communications 9, 2385 (2018)

11 Nature Electronics 2, 115 (2019)

12 Nature communications 9, 417 (2018)