Patent US5121228A successfully refreshed from Google Patents.
US5121228A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Holographic learning machine
Assignee: Leland Stanford Junior University
•
Filing: 1990-09-27
•
Granted: 1992-06-09
•
Expires: 2010-09-27
Abstract
Abstract A holographic learning machine comprising a volume holographic recording medium, 2N photo-detectors receiving light diffracted from the medium, and N electrical circuits subtracting respective pairs of outputs of the photo-detectors, thresholding the differences and comparing them to target values. During training, a coherent image beam shines on the recording medium and corresponding target values are supplied to the circuits. Depending on whether the comparison between the thresholded difference and the target is positive or negative, a positive error or negative error amplitude modulator in a 2N array is turned on to transmit to the recording medium light coherent with the image beam. Thereby, the full bipolar perceptron algorithm is implemented. The implementation may be integrated into a rugged, integral unit.
Inventors
- 👤 Eung-Gi Paek
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-01 11:35
CPC Classifications
G; G03; G03H; G03H1/00; G03H1/02; G03H1/024; G03H1/0248; G03H1/04; G03H1/16; G09; G09B; G09B5/00; G09B5/02; G03H1/0005; G03H1/26; G03H1/2645; G03H1/265; G03H2260/00; G03H2260/50; G03H2260/54
Claims Analysis
Total Claims: 13
Claim 1
Independent Claim
170 words
A holographic learning machine, comprising: a holographic recording medium; a source of a first beam and a second beam of light coherent with each other; a source of a set of training vectors impressed on a spatial extent of said first beam of light; a first optical system distributing said first beam through said recording medium; an array of 2N detectors, N being greater than 1; a second optical system distributing a third beam diffracted by said recording medium from said first beam onto said detectors; N processing units, each receiving and combining inputs from at least two of said detectors and a target signal corresponding to a selected one of said set of training vectors and producing a positive-error signal and a negative-error signal, 2N error signals being thereby produced; an array of 2N optical modulators simultaneously spatially modulating said second beam and receiving respective ones of said error signals; and a third optical system distributing said second beam modulated by said array of modulators onto said recording medium.
Claim 2
Dependent Claim
170 words
A holographic learning machine, comprising: a holographic recording medium; a source of a first beam and a second beam of light coherent with each other; a source of a set of training vectors impressed on a spatial extent of said first beam of light; a first optical system distributing said first beam through said recording medium; an array of 2N detectors, N being greater than 1; a second optical system distributing a third beam diffracted by said recording medium from said first beam onto said detectors; N processing units, each receiving and combining inputs from at least two of said detectors and a target signal corresponding to a selected one of said set of training vectors and producing a positive-error signal and a negative-error signal, 2N error signals being thereby produced; an array of 2N optical modulators simultaneously spatially modulating said second beam and receiving respective ones of said error signals; and a third optical system distributing said second beam modulated by said array of modulators onto said recording medium.
Claim 3
Dependent Claim
47 words
A holographic learning machine as recited in claim 1, wherein each of said processing units includes: a recognition circuit comparing outputs of two of said detectors and thresholding a result of said comparison; and an error circuit comparing a result of said thresholding with said target signal.
Claim 4
Dependent Claim
24 words
A holographic learning machine as recited in claim 1, wherein each of said first, second, and third optical systems comprises at least one lens.
Claim 5
Dependent Claim
27 words
A holographic learning machine as recited in claim 1, wherein each of said first and third optical systems comprises a diffusive medium attached to said recording medium.
Claim 6
Dependent Claim
22 words
A holographic learning machine as recited in claim 4, wherein said second optical system comprises a lens attached to said recording medium.
Claim 7
Dependent Claim
43 words
A holographic learning machine as recited in claim 5, wherein said array of optical modulators are attached to said recording medium and wherein said array of detectors is attached to a medium surrounding a periphery of said lens of said second optical system.
Claim 8
Dependent Claim
22 words
A holographic learning machine as recited in claim 6, wherein said processing units are attached to a surface of said recording medium.
Claim 9
Dependent Claim
20 words
A holographic learning machine as recited in claim 1, wherein each of said processing units is an electrical processing unit.
Claim 10
Dependent Claim
37 words
A holographic learning machine as recited in claim 1, wherein each of said processing units implements a complete perceptron algorithm acting upon said received inputs and said received target signal and providing said positive-error and negative-error signals.
Claim 11
Dependent Claim
26 words
A holographic learning machine as recited in claim 1, wherein said training vectors are 2-dimensional images impressed on an area of said first beam of light.
Claim 12
Dependent Claim
55 words
An integrated holographic learning machine, comprising: a body of a recordable photorefractive material; a two-dimensional selectively transmissive array attached to a face of said body; a linear selectively transmissive array attached to said face; a lens rigidly attached to said body; and a linear detector array rigidly attached to said body with said lens therebetween.
Claim 13
Dependent Claim
55 words
An integrated holographic learning machine, comprising: a body of a recordable photorefractive material; a two-dimensional selectively transmissive array attached to a face of said body; a linear selectively transmissive array attached to said face; a lens rigidly attached to said body; and a linear detector array rigidly attached to said body with said lens therebetween.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1995-06-09 | 1995-12-09 | 1996-06-09 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1999-06-09 | 1999-12-09 | 2000-06-09 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2003-06-09 | 2003-12-09 | 2004-06-09 | $7700.00 | Window Closed / Lapsed |