Adaptive, optical, radial basis function neural network for handwritten digit recognition

Foor, Wesley E.; Neifeld, Mark A.

doi:10.1364/ao.34.007545

Cited by 5 publications

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Continued technology developments in optical and optoelectronic information processing in combination with mature VLSI technology hold the potential for significant performance improvements in artificial neural information processing systems [151]- [159], promising massive inter-chip connectivity as needed for larger size neural networks. High-density optical storage and adaptation for integrated learning could be achieved in 3-D optical media such as photorefractive crystals.…”

Section: Emerging Technologiesmentioning

confidence: 99%

Neuromorphic Learning VLSI Systems: A Survey

Cauwenberghs¹

The Springer International Series in Engineering and Computer Science

View full text Add to dashboard Cite

INTRODUCTIONCarver Mead introduced "neuromorphic engineering" [1] as an interdisciplinary approach to the design of biologically inspired neural information processing systems, whereby neurophysiological models of perception and information processing in biological systems are mapped onto analog VLSI systems that not only emulate their functions but also resemble their structure [18]. The motivation for emulating neural function and structure in analog VLSI is the realization that challenging tasks of perception, classification, association and control successfully performed by living organisms can only be accomplished in artificial systems by using an implementation medium that matches their structure and organization.Essential to neuromorphic systems are mechanisms of adaptation and learning, modeled after the "plasticity" of synapses and neural structure in biological systems [25,4]. Learning can be broadly defined as a special case of adaptation whereby past experience is used effectively in readjusting the system response to previously unseen, although similar, stimuli. Based on the nature and availability of a training feedback signal, learning algorithms for artificial neural networks fall under three broad categories: unsupervised, supervised and reward/punishment (reinforcement). Physiological experiments have revealed plasticity mechanisms in biology that correspond to Hebbian unsupervised learning [19], and classical (pavlovian) conditioning [17,22] characteristic of reinforcement learning.

show abstract

Section: Emerging Technologiesmentioning

confidence: 99%