Project Info
Memory elements for Neuromorphic Computing using Superconductor Ferroelectric Hybrids
Geoff Brennecka | gbrennec@mines.edu and Meenakshi Singh | msingh@mines.edu
As the limits continue to be pushed on conventional CMOS von Neumann computer architectures, increasing attention is being given to computation by alternative (non von Neumann) models such as neuromorphic architectures. Memristive devices form the basis of many neuromorphic applications. Superconducting-ferroelectric hybrids might be particularly promising as memristive elements for neuromorphic computing because of their low energy consumption and fast switching speeds. Therefore, they are interesting systems to study from this perspective.
Elucidate the interdisciplinary nature of the project
The project involves material fabrication (mentored by GB in MME) and device fabrication and measurement (mentored by MS in Physics).
More Information
I. K. Schuller and R. Stevens, Neuromorphic Computing : From Materials to Systems Architecture Report of a Roundtable Convened to Consider Neuromorphic Computing. 2015.
S. Boyn et al., “Learning through ferroelectric domain dynamics in solid-state synapses,” Nat. Commun., vol. 8, p. 14736, 2017.
Grand Engineering Challenge: Engineer the tools of scientific discovery
Student Preparation
Qualifications
1. Experience with experimental projects or lab courses a must
2. Proficiency in Python not required but beneficial
3. Some exposure to solid state Physics not required but beneficial
Time Commitment
5 hours/week
Skills/Techniques Gained
1. clean room experience
2. thin film deposition
3. wire-bonding
4. cryogenic measurement
5. AC measurement techniques (maybe microwave)
6. programming for data acquisition
Mentoring Plan
Individual meetings every week with one of the mentors. Group meetings with advisor’s research group once every two weeks. Group meeting of this particular project group once in 5 weeks. In addition, hands on training in the lab with advisor and senior graduate students.