Bio-Inspired Stochastic Computing Using Binary CBRAM Synapses

In this paper, we present an alternative approach to neuromorphic systems based on multilevel resistive memory synapses and deterministic learning rules. We demonstrate an original methodology to use conductive-bridge RAM (CBRAM) devices as, easy to program and low-power, binary synapses with stocha...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on electron devices 2013-07, Vol.60 (7), p.2402-2409
Main Authors: Suri, M., Querlioz, D., Bichler, O., Palma, G., Vianello, E., Vuillaume, D., Gamrat, C., DeSalvo, B.
Format: Article
Language:English
Subjects:
Applied sciences
Auditory learning
CBRAM synapse
Design. Technologies. Operation analysis. Testing
Electronics
Engineering Sciences
Exact sciences and technology
Immune system
Integrated circuits
Integrated circuits by function (including memories and processors)
Mathematical model
Neurons
Probabilistic logic
Programming
Semiconductor electronics. Microelectronics. Optoelectronics. Solid state devices
spike-timing dependent plasticity (STDP)
stochastic neuromorphic system
Switches
visual pattern extraction
Visualization
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we present an alternative approach to neuromorphic systems based on multilevel resistive memory synapses and deterministic learning rules. We demonstrate an original methodology to use conductive-bridge RAM (CBRAM) devices as, easy to program and low-power, binary synapses with stochastic learning rules. New circuit architecture, programming strategy, and probabilistic spike-timing dependent plasticity (STDP) learning rule for two different CBRAM configurations with-selector (1T-1R) and without-selector (1R) are proposed. We show two methods (intrinsic and extrinsic) for implementing probabilistic STDP rules. Fully unsupervised learning with binary synapses is illustrated through two example applications: 1) real-time auditory pattern extraction (inspired from a 64-channel silicon cochlea emulator); and 2) visual pattern extraction (inspired from the processing inside visual cortex). High accuracy (audio pattern sensitivity > 2, video detection rate > 95%) and low synaptic-power dissipation (audio 0.55 μW, video 74.2 μW) are shown. The robustness and impact of synaptic parameter variability on system performance are also analyzed.
ISSN:0018-9383
1557-9646
DOI:10.1109/TED.2013.2263000