Award Abstract #0103059
Nanoscale Single-electron Switching Arrays for Self-evolving Neuromorphic Networks

NSF Org:	CCF Division of Computer and Communication Foundations
Initial Amendment Date:	July 23, 2001
Latest Amendment Date:	July 23, 2001
Award Number:	0103059
Award Instrument:	Standard Grant
Program Manager:	S. Kamal Abdali CCF Division of Computer and Communication Foundations CSE Directorate for Computer & Information Science & Engineering
Start Date:	July 1, 2001
Expires:	December 31, 2003 (Estimated)
Awarded Amount to Date:	$599980
Investigator(s):	Konstantin Likharev klikharev@notes.cc.sunysb.edu (Principal Investigator) Andreas Mayr (Co-Principal Investigator) Michael Bender (Co-Principal Investigator)
Sponsor:	SUNY at Stony Brook WEST 5510 FRK MEL LIB STONY BROOK, NY 11794 631/632-9949
NSF Program(s):	SPECIAL PROJECTS - CCF, ENGINEERING RESEARCH CENTERS, PARTICULATE &MULTIPHASE PROCES
Field Application(s):
Program Reference Code(s):	HPCC, 9216, 1674
Program Element Code(s):	2878, 1480, 1415

ABSTRACT

The goal of this project is to carry out a detailed multi-disciplinary study of single-electron

latching switches and of possible use of 2D arrays of such switches for hardware

implementation of self-organizing (plastic) neuromorphic networks. Preliminary estimates

show that such networks may provide unparalleled possibilities for complex information

processing. By these estimates, the networks may also have remarkable scaling properties:

if implemented using a 10-nm technology, they may have density about 10 8 neurons per

cm 2 at manageable power dissipation below 100 W/cm 2 , and feature full learning cycle time

of the order of a few seconds. This scaling gives every hope that the networks will be able,

after initial (largely unsupervised) learning, not only provide complex information processing

including complex image recognition, but possibly reproduce biological evolution of the

cerebral cortex at a time scale some 6 orders of magnitude shorter.

The objective of the proposed project is to carry out a preliminary study of this

remarkable opportunity, addressing all its basic aspects at several structural levels. In

particular, research will include the following components:

A. Single-electron switch node design (D. Averin, K. Likharev, J. Wells).

Detailed theoretical analysis and modeling (on two basic levels of single-electron transport

theory) of statics, dynamics, and statistics of the proposed single-electron latching switches.

B. Low temperature prototyping (J. Lukens). Fabrication and experimental

study of Al/AlOx/Al prototypes of single-electron latching switches, with the goal to scale

single-electron islands down to 100 nm and tunnel junctions to 10 nm, respectively, which

would bring the reliable operation temperature up to about 10 K.

C. Molecular single-electron device development (B. Brunschwig, J. Lukens,

A. Mayr). Exploration of the opportunity to implement the basic component of the switches,

the single-electron transistor, by chemical self-assembly of molecular components. The

molecular components will be deposited in solution on the prefabricated metallic wire

structures, and then characterized using a set of electrical, electrochemical, and time-resolved

laser-spectrometry methods.

D. Top level modeling and analysis (J. Barhen, M. Bender, K. Likharev).

Large-scale computer simulation and a partial analytical study of the growth, dynamics, and

self-adaptation of neuromorphic networks based on these switches.

Hopefully, the project will achieve enough progress to justify a large-scale R&D effort

in this exciting direction. In particular, a reliable evidence of self-organization of adaptive

neuromorphic networks during largely unsupervised learning would certainly be followed by

the first hardware implementations of sizable networks (possibly, after an initial stage of

purely-CMOS-based prototyping using commercially available FPGA technology).

The project will have a substantial educational component. Specifically (besides

participating in general educational Stony Brook initiatives), at least 4 FTE graduate

students will be involved in the project each year, and some 20 undergraduate and

graduate students will take part in the project during its full 4-year period. At least one

student will work in BNL and one in ORNL most of the time. Working in a multi-disciplinary

team will allow these students to overcome inter-departmental barriers in their education.

As another specific educational initiative, we plan to organize a Web-based undergraduate

course on massively parallel supercomputing and neural networks, using the IBM SP3

computer at Oak Ridge.

Work on the inter-related aspects of this multi-disciplinary project will be constantly

coordinated by its P.I. (K. Likharev). In particular, regular meetings of all Stony Brook and

Brookhaven participants of the team working on the project (including postdoctoral

associates and students), and annual meetings with Oak Ridge collaborators, are planned.

Please report errors in award information by writing to: awardsearch@nsf.gov.