Perceptrons with Hebbian Learning Based on Wave Ensembles in Spatially Patterned Potentials
Authors:
- 18 March 2015
- journal article
- research article
- Published by American Physical Society (APS) in Physical Review Letters
- Vol. 114 (11) , 118101
Abstract
A general scheme to realize a perceptron for hardware neural networks is presented, where multiple interconnections are achieved by a superposition of Schrödinger waves. Spatially patterned potentials process information by coupling different points of reciprocal space. The necessary potential shape is obtained from the Hebbian learning rule, either through exact calculation or construction from a superposition of known optical inputs. This allows implementation in a wide range of compact optical systems, including (1) any nonlinear optical system, (2) optical systems patterned by optical lithography, and (3) exciton-polariton systems with phonon or nuclear spin interactions.Keywords
All Related Versions
Funding Information
- Nantong University (M4080853.110.706022)
- Lee Kuan Yew Endowment Fund
This publication has 36 references indexed in Scilit:
- Spin-torque building blocksNature Materials, 2013
- Neural networks using two-component Bose-Einstein condensatesScientific Reports, 2013
- Principles of Artificial Neural Networks2013
- The Memristive Magnetic Tunnel Junction as a Nanoscopic Synapse‐Neuron SystemAdvanced Materials, 2012
- Short-term plasticity and long-term potentiation mimicked in single inorganic synapsesNature Materials, 2011
- Neuromorphic Silicon Neurons and Large-Scale Neural Networks: Challenges and OpportunitiesFrontiers in Neuroscience, 2011
- Neuromorphic Silicon Neuron CircuitsFrontiers in Neuroscience, 2011
- Artificial neural networks in hardware: A survey of two decades of progressNeurocomputing, 2010
- Neural Networks1996
- Optical neurochip based on a three-layered feed-forward modelOptics Letters, 1990