A Mixed-Signal Spiking Neuromorphic Architecture for Scalable Neural Network

Amorphous Computational Material (ACM) is a concept of an active material that can sense its environment and, due to its cognitive capabilities, react “intelligently” to those changes. In this paper, We demonstrate the feasibility of utilizing water hammer as a form of directed actuation. We show a novel concept of a Synthetic Neural Network, a type of an organic neuromorphic architecture modeled after Artificial Neural Network, which is used for a distributed cognition purposes for ACM. A simulation of the SNN is shown to accurately predict the directionality of water hammer propulsion.

Download Full-text

A neuromorphic systems approach to in-memory computing with non-ideal memristive devices: from mitigation to exploitation

Faraday Discussions ◽

10.1039/c8fd00114f ◽

2019 ◽

Vol 213 ◽

pp. 487-510 ◽

Cited By ~ 16

Author(s):

Melika Payvand ◽

Manu V. Nair ◽

Lorenz K. Müller ◽

Giacomo Indiveri

Keyword(s):

Neural Network ◽

Network Architecture ◽

Systems Approach ◽

Spiking Neural Network ◽

Neural Network Architecture ◽

Mixed Signal ◽

Neuromorphic Systems

In this paper, we present a spiking neural network architecture that supports the use of non-ideal memristive devices as synaptic elements and propose mixed-signal analog-digital interfacing circuits to mitigate/exploit such non-idealities for neuromorphic computation.

Download Full-text

A mixed-signal VLSI neural network with on-chip learning

CCECE 2003 - Canadian Conference on Electrical and Computer Engineering. Toward a Caring and Humane Technology (Cat. No.03CH37436) ◽

10.1109/ccece.2003.1226465 ◽

2004 ◽

Author(s):

M. Mirhassani ◽

M. Ahmadi ◽

W.C. Miller

Keyword(s):

Neural Network ◽

Mixed Signal ◽

On Chip

Download Full-text

The simplicial neural cell and its mixed-signal circuit implementation: an efficient neural-network architecture for intelligent signal processing in portable multimedia applications

IEEE Transactions on Neural Networks ◽

10.1109/tnn.2002.1021899 ◽

2002 ◽

Vol 13 (4) ◽

pp. 995-1008 ◽

Cited By ~ 28

Author(s):

R. Dogaru ◽

P. Julian ◽

L.O. Chua ◽

M. Glesner

Keyword(s):

Neural Network ◽

Signal Processing ◽

Network Architecture ◽

Neural Cell ◽

Multimedia Applications ◽

Circuit Implementation ◽

Neural Network Architecture ◽

Mixed Signal

Download Full-text

Mapping Hebbian Learning Rules to Coupling Resistances for Oscillatory Neural Networks

Frontiers in Neuroscience ◽

10.3389/fnins.2021.694549 ◽

2021 ◽

Vol 15 ◽

Author(s):

Corentin Delacour ◽

Aida Todri-Sanial

Keyword(s):

Neural Network ◽

Large Scale ◽

Hebbian Learning ◽

Hopfield Neural Network ◽

Analytical Framework ◽

Relaxation Oscillators ◽

Oscillatory Neural Network ◽

Network Functionality ◽

Fully Connected ◽

Neuromorphic Architecture

Oscillatory Neural Network (ONN) is an emerging neuromorphic architecture with oscillators representing neurons and information encoded in oscillator's phase relations. In an ONN, oscillators are coupled with electrical elements to define the network's weights and achieve massive parallel computation. As the weights preserve the network functionality, mapping weights to coupling elements plays a crucial role in ONN performance. In this work, we investigate relaxation oscillators based on VO2 material, and we propose a methodology to map Hebbian coefficients to ONN coupling resistances, allowing a large-scale ONN design. We develop an analytical framework to map weight coefficients into coupling resistor values to analyze ONN architecture performance. We report on an ONN with 60 fully-connected oscillators that perform pattern recognition as a Hopfield Neural Network.

Download Full-text