1986
DOI: 10.1364/ol.11.000482
|View full text |Cite
|
Sign up to set email alerts
|

Optical implementation of associative memory with controlled nonlinearity in the correlation domain

Abstract: A mathematical model for incorporating controllable nonlinearity in the correlation domain of a conventional associative memory is described. Such a mechanism provides the flexibility of rapidly and arbitrarily changing the strengths of the stored states in an associative memory. Such a feature corresponds to shifting of attention in psychological terms. This attentive associative memory can be implemented optically. Results obtained with computer simulation and a design for a compact optical implementation ar… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
12
0

Year Published

1988
1988
2022
2022

Publication Types

Select...
6
1
1

Relationship

1
7

Authors

Journals

citations
Cited by 83 publications
(12 citation statements)
references
References 2 publications
0
12
0
Order By: Relevance
“…Fukushima h developed neural network model (selective attention) for character segmentation in \ Neocognitron [Fukushima (1987)]. The attentive associative memory model implement opto-electronically by Athale, Szu & Friedlander (1986) can be augmented by a priori probabili compiled by a character-pair correlation function of connected characters. This is an interesti: area for more research.…”
Section: Resultsmentioning
confidence: 99%
“…Fukushima h developed neural network model (selective attention) for character segmentation in \ Neocognitron [Fukushima (1987)]. The attentive associative memory model implement opto-electronically by Athale, Szu & Friedlander (1986) can be augmented by a priori probabili compiled by a character-pair correlation function of connected characters. This is an interesti: area for more research.…”
Section: Resultsmentioning
confidence: 99%
“…The second hologram is prepared by a similar procedure except that the associated output vectors are recorded in correspondence to each point in the two dimensional training plane. After the holograms are thus prepared, an input vector is loaded into the onedimcasional input array and the correlations between it and the M m,-.mory vectors are displayed in the OUtput plane (Athale, Szu, & Friedlander, 1986;Owechko, Dunning, Marom, & Softer, 1987;Paek & Psalfis, 1987). An optically addressed SLM can be used to produce an amplitude distribution which is the square of the incident ~ amplitudes.…”
Section: Desired Outputsmentioning
confidence: 99%
“…Recently considerable efforts were made to implement neural-network models [2][3][4] in optics . As a result, a number of attempts at designing an optical AM based on these models have been successful [5][6][7][8][9][10][11][12][13] . These two approaches to AM implementation have much in common .…”
Section: Introductionmentioning
confidence: 99%
“…Then, in a time interval less than the medium relaxation time a third optical pulse 8 3 (x, z, t) is fed to excite the sample . This pulse has a central optical frequency co and is resonant for transition from the ground state In> to the excitation one In,> .…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation