US4967369A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Variable capacity transmission medium
Assignee: Individual
•
Filing: 1988-06-07
•
Granted: 1990-10-30
•
Expires: 2008-06-07
Abstract
Abstract Methods of information correlation based on extremization of transmission capacity of an assembly of information processors are disclosed. In one embodiment, an assembly of processors exchanging information through the exchange of random codes serves as a correlator between two inputs adjusting its total communication capacity to achieve maximum correlation between two inputs.
Inventors
- 👤 Alexander N. Jourjine
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-29 16:50
CPC Classifications
G; G06; G06N; G06N3/00; G06N3/02; G06N3/04; G06N3/044; G06N3/0442
Claims Analysis
Total Claims: 8
Claim 1
Independent Claim
135 words
A method of weight adjustment in a neural network comprising an assembly of input/output processors with information transmission cells, the method comprising: providing inputs for activations of input/output units arranged on opposite sides of a two dimensional array of information transmission cells; making said information transmission cells communicate with each other by emission and absorption of signals modulated according to random binary codes; arranging said codes to be non-zero within activation windows of a pulse length, the pulse length being within a longer code length; making said information transmission cells maximize their transmission capacity by maximizing heat emission from said information transmission cells; maximizing said heat emission by updating position of said activation windows within said longer code lengths; and measuring the total heat flux by said assembly of input/output processors and information transmission cells.
Claim 2
Independent Claim
135 words
A method of weight adjustment in a neural network comprising an assembly of input/output processors with information transmission cells, the method comprising: providing inputs for activations of input/output units arranged on opposite sides of a two dimensional array of information transmission cells; making said information transmission cells communicate with each other by emission and absorption of signals modulated according to random binary codes; arranging said codes to be non-zero within activation windows of a pulse length, the pulse length being within a longer code length; making said information transmission cells maximize their transmission capacity by maximizing heat emission from said information transmission cells; maximizing said heat emission by updating position of said activation windows within said longer code lengths; and measuring the total heat flux by said assembly of input/output processors and information transmission cells.
Claim 3
Dependent Claim
175 words
A random code neural network for information correlation and discrimination of plural inputs comprising: an array of information transmission cells for receiving the plural inputs, each information transmission cell comprising a transmission cell and a memory cell, the information transmission cells communicating with each other by emission and absorption of signals modulated according to random binary codes, the binary codes from each information transmission cell being non-zero within activation windows of a pulse length, the pulse length being within a longer code length; each transmission cell comprising means for measuring energy flux in an input and means for learning by updating timings of random code to increase the transmission capacity of the random code; each memory cell comprising means for measuring energy flux in an input and means for learning by updating timings of random code at a speed which is low relative to the speed of updating timings in the transmission cell to increase the transmission capacity of the random code, and means for monitoring the learning of the array of information transmission cells.
Claim 4
Dependent Claim
175 words
A random code neural network for information correlation and discrimination of plural inputs comprising: an array of information transmission cells for receiving the plural inputs, each information transmission cell comprising a transmission cell and a memory cell, the information transmission cells communicating with each other by emission and absorption of signals modulated according to random binary codes, the binary codes from each information transmission cell being non-zero within activation windows of a pulse length, the pulse length being within a longer code length; each transmission cell comprising means for measuring energy flux in an input and means for learning by updating timings of random code to increase the transmission capacity of the random code; each memory cell comprising means for measuring energy flux in an input and means for learning by updating timings of random code at a speed which is low relative to the speed of updating timings in the transmission cell to increase the transmission capacity of the random code, and means for monitoring the learning of the array of information transmission cells.
Claim 5
Dependent Claim
23 words
A random code neural network as claimed in claim 2 wherein the means for monitoring measures the total transmission capacity of the array.
Claim 6
Dependent Claim
20 words
A random code neural network as claimed in claim 2 wherein the means for monitoring comprises a heat flux detector.
Claim 7
Dependent Claim
27 words
A random code neural network as claimed in claim 2 further comprising a common random code generator which supplies a random code to each information transmission cell.
Claim 8
Dependent Claim
33 words
A random code neural network as claimed in claim 2 wherein the means for updating timings to increase the transmission capacity of the random code maximizes heat emission from the information transmission cell.
References Cited (Backward Citations)
US4660166A
Cited
US4747069A
Cited
US4862406A
Cited
US4802103A
Cited
US4807168A
Cited
US4858147A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1993-10-29 | 1994-04-30 | 1994-10-30 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1997-10-29 | 1998-04-30 | 1998-10-30 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2001-10-29 | 2002-04-30 | 2002-10-30 | $7700.00 | Window Closed / Lapsed |