CaltechAUTHORS
  A Caltech Library Service

Synthesizing cognition in neuromorphic electronic systems

Neftci, Emri and Binas, Jonathan and Rutishauser, Ueli and Chicca, Elisabetta (2013) Synthesizing cognition in neuromorphic electronic systems. Proceedings of the National Academy of Sciences of the United States of America, 110 (37). E3468-E3476. ISSN 0027-8424. PMCID PMC3773754. https://resolver.caltech.edu/CaltechAUTHORS:20150126-105424388

[img] PDF - Published Version
See Usage Policy.

1225Kb
[img] PDF - Supplemental Material
See Usage Policy.

712Kb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20150126-105424388

Abstract

The quest to implement intelligent processing in electronic neuromorphic systems lacks methods for achieving reliable behavioral dynamics on substrates of inherently imprecise and noisy neurons. Here we report a solution to this problem that involves first mapping an unreliable hardware layer of spiking silicon neurons into an abstract computational layer composed of generic reliable subnetworks of model neurons and then composing the target behavioral dynamics as a “soft state machine” running on these reliable subnets. In the first step, the neural networks of the abstract layer are realized on the hardware substrate by mapping the neuron circuit bias voltages to the model parameters. This mapping is obtained by an automatic method in which the electronic circuit biases are calibrated against the model parameters by a series of population activity measurements. The abstract computational layer is formed by configuring neural networks as generic soft winner-take-all subnetworks that provide reliable processing by virtue of their active gain, signal restoration, and multistability. The necessary states and transitions of the desired high-level behavior are then easily embedded in the computational layer by introducing only sparse connections between some neurons of the various subnets. We demonstrate this synthesis method for a neuromorphic sensory agent that performs real-time context-dependent classification of motion patterns observed by a silicon retina.


Item Type:Article
Related URLs:
URLURL TypeDescription
http://dx.doi.org/10.1073/pnas.1212083110DOIArticle
http://www.pnas.org/lookup/suppl/doi:10.1073/pnas.1212083110/-/DCSupplementalPublisherSupporting Information
http://ncs.ethz.ch/projects/vlsi-wta-networks/synthesizing-cognition-in-neuromorphic-vlsi-systems-experimental-data/viewRelated ItemExperimental Data
http://ncs.ethz.ch/projects/vlsi-wta-networks/synthesizing-cognition-in-neuromorphic-vlsi-systems-scripts/viewRelated ItemSimulation Scripts
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3773754/PubMed CentralArticle
ORCID:
AuthorORCID
Rutishauser, Ueli0000-0002-9207-7069
Contact Email Address:urut@caltech.edu
Additional Information:© 2013 National Academy of Sciences. Edited by Terrence J. Sejnowski, Salk Institute for Biological Studies, La Jolla, CA, and approved June 10, 2013 (received for review July 20, 2012). Published online before print July 22, 2013, doi: 10.1073/pnas.1212083110 We thank Daniel Fasnacht for designing the AER infrastructure; Chiara Bartolozzi for the SAC; Tobi Delbruck for discussion and jAER support; Tobi Delbruck and Patrick Lichtsteiner for the DVS camera; Shih-Chii Liu, Jean-Jacques Slotine, and Matthew Cook for discussion; and Michael Pfeiffer and Florian Jug for review. This work was supported by the European Union (EU) European Research Council Grant “neuroP” (257219), by the EU Information and Communication Technologies Grant “acoustic SCene ANalysis for Detecting Living Entities (SCANDLE)” (231168), and by the Excellence Cluster 227 (Cognitive Interaction Technology–Center of Excellence, Bielefeld University). Author contributions: E.N., J.B., U.R., E.C., G.I., and R.J.D. designed research; E.N. and J.B. performed research; E.N. and J.B. contributed new reagents/analytic tools; E.N. and J.B. analyzed data; and E.N., J.B., U.R., E.C., G.I., and R.J.D. wrote the paper. The authors declare no conflict of interest. This Direct Submission article had a prearranged editor. Data deposition: The experimental data can be found under http://ncs.ethz.ch/projects/vlsi-wta-networks/synthesizing-cognition-in-neuromorphic-vlsi-systems-experimental-data/view. Simulations scripts can be found under http://ncs.ethz.ch/projects/vlsi-wta-networks/synthesizing-cognition-in-neuromorphic-vlsi-systems-scripts/view. This website is provided by Eidgenössiche Technische Hochschule (ETH) Zurich and maintained by the Neuromorphic Cognitive Systems (NCS) group at the institute of Neuroinformatics, University of Zurich and ETH Zurich. This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1212083110/-/DCSupplemental.
Funders:
Funding AgencyGrant Number
European Research Council (ERC)“neuroP” (257219)
European Research Council (ERC)“acoustic SCene ANalysis for Detecting Living Entities (SCANDLE)” (231168)
Bielefeld UniversityUNSPECIFIED
Subject Keywords:decision making; sensorimotor; working memory; analog very large-scale integration; artificial neural systems
Issue or Number:37
PubMed Central ID:PMC3773754
Record Number:CaltechAUTHORS:20150126-105424388
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20150126-105424388
Official Citation:Emre Neftci, Jonathan Binas, Ueli Rutishauser, Elisabetta Chicca, Giacomo Indiveri, and Rodney J. Douglas Synthesizing cognition in neuromorphic electronic systems PNAS 2013 110 (37) E3468-E3476; published ahead of print July 22, 2013, doi:10.1073/pnas.1212083110
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:54077
Collection:CaltechAUTHORS
Deposited By: SWORD User
Deposited On:26 Jan 2015 20:21
Last Modified:01 Apr 2020 22:48

Repository Staff Only: item control page