US5113483A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Neural network with semi-localized non-linear mapping of the input space
Assignee: Rateze Remote Mgmt LLC
•
Filing: 1990-06-15
•
Granted: 1992-05-12
•
Expires: 2010-06-15
Abstract
Abstract A neural network includes an input layer comprising a plurality of input units (24) interconnected to a hidden layer with a plurality of hidden units (26) disposed therein through an interconnection matrix (28). Each of the hidden units (26) is a single output that is connected to output units (32) in an output layer through an interconnection matrix (30). Each of the interconnections between one of the hidden units (26) to one of the output units (32) has a weight associated therewith. Each of the hidden units (26) has an activation in the i'th dimension and extending across all the other dimensions in a non-localized manner in accordance with the following equation: <IMAGE> that the network learns by the Back Propagation method to vary the output weights and the parameters of the activation function mu hi and sigma hi.
Inventors
- 👤 James D. Keeler
- 👤 Eric J. Hartman
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-26 00:13
CPC Classifications
G; G06; G06N; G06N3/00; G06N3/02; G06N3/04; G06N3/0495; G06N3/0499; G06N3/08; G06N3/09; G06V; G06V10/00; G06V10/40; G06V10/44; G06V10/443; G06V10/449; G06V10/451; G06V10/454
Claims Analysis
Total Claims: 19
Claim 1
Independent Claim
266 words
A neural network for effecting non-linear mapping of an input space, comprising: an input layer for receiving a multi-dimensional input vector x i , for i ranging from 1 to n, n being an integer; a hidden layer, having m hidden units, m being an integer, each of said hidden units receiving a plurality of inputs and providing a single output and having an activation function: ##EQU12## where: h is the number of the associated one of said hidden units, μ hi is the center of a h , f hi is a localized function in the x i dimension and non-localized in at least one other of said x i dimensions, C hi is a constant and {P hi } is a set of parameters for f hi ; a connection matrix for interconnecting the output of select ones of the input vectors x i to the inputs of said hidden units in accordance with an interconnection scheme; an output layer comprising M output units, M being an integer, each of said output units for providing a single output Y i for i ranging from 1 to M, each of said output units having a predetermined transfer function g for i ranging from 1 to M; and an output interconnection matrix for interconnecting the output of select ones of said hidden units to the inputs of select ones of said M output units in accordance with an interconnection scheme, each of said interconnections between the output of said hidden units and the associated one of said M output units having an output weight associated therewith.
Claim 2
Dependent Claim
266 words
A neural network for effecting non-linear mapping of an input space, comprising: an input layer for receiving a multi-dimensional input vector x i , for i ranging from 1 to n, n being an integer; a hidden layer, having m hidden units, m being an integer, each of said hidden units receiving a plurality of inputs and providing a single output and having an activation function: ##EQU12## where: h is the number of the associated one of said hidden units, μ hi is the center of a h , f hi is a localized function in the x i dimension and non-localized in at least one other of said x i dimensions, C hi is a constant and {P hi } is a set of parameters for f hi ; a connection matrix for interconnecting the output of select ones of the input vectors x i to the inputs of said hidden units in accordance with an interconnection scheme; an output layer comprising M output units, M being an integer, each of said output units for providing a single output Y i for i ranging from 1 to M, each of said output units having a predetermined transfer function g for i ranging from 1 to M; and an output interconnection matrix for interconnecting the output of select ones of said hidden units to the inputs of select ones of said M output units in accordance with an interconnection scheme, each of said interconnections between the output of said hidden units and the associated one of said M output units having an output weight associated therewith.
Claim 3
Dependent Claim
26 words
The neural network of claim 1 wherein said activation function a h is Gaussian according to the following equation: ##EQU13## where: σ hi =the standard deviation.
Claim 4
Dependent Claim
37 words
The neural network of claim 1 wherein the constant C hi comprises input weights associated with each interconnection between one of said input vectors x i and the interconnected one of the inputs of said hidden units.
Claim 5
Dependent Claim
22 words
The neural network of claim 1 wherein said predetermined transfer function g i in said m output units is a sigmoidal function.
Claim 6
Dependent Claim
12 words
The neural network of claim 1 wherein said output weights are variable.
Claim 7
Dependent Claim
65 words
The neural network of claim 5 and further comprising learning means for varying the value of μ hi in said activation function and said output weights in response to the input of known values for said input vectors and a predetermined target output such that said target output is realized after a finite number of said known input values are input to said input layer.
Claim 8
Dependent Claim
16 words
The neural network of claim 6 wherein said learning algorithm is a Back Propagation learning algorithm.
Claim 9
Dependent Claim
23 words
The neural network of claim 6 wherein said means for minimizing the error minimizes the error in accordance with a competitive learning algorithm.
Claim 10
Dependent Claim
62 words
The neural network of claim 1 wherein the output of each of said output units is defined by the following relationship: ##EQU14## where: i ranges from 1 to M h is the number of said hidden unit from 1 to m, and w hi is the value of said output weight on the output of the h'th one of said hidden units.
Claim 11
Independent Claim
235 words
A method for mapping an input space in the neural network, comprising the steps of: receiving a multi-dimensional input vector x i , for i ranging from 1 to n, n being an integer; providing m hidden units, m being an integer, each of the hidden units providing a single output value and having an activation function: ##EQU15## where: h is the number of the associated one of the hidden units, μ hi is the localized center of a h in the x 1 dimension, and f hi is a localized function in the x i dimension and non-localized in at least one other of the x i dimensions; interconnecting the output of select ones of the input vectors x i to the inputs of the hidden units in accordance with an interconnection scheme; providing a plurality of output units in an output layer, each of the output units for providing a single output y i for i ranging from 1 to M, M being an integer, each of the output units having associated therewith a predetermined control function g; and interconnecting the output of select ones of the hidden units to the inputs of predetermined ones of the M output units in accordance with the predetermined interconnection scheme, each of the interconnections between the output of the hidden units and the associated one of the M output units having an output weight associated therewith.
Claim 12
Independent Claim
235 words
A method for mapping an input space in the neural network, comprising the steps of: receiving a multi-dimensional input vector x i , for i ranging from 1 to n, n being an integer; providing m hidden units, m being an integer, each of the hidden units providing a single output value and having an activation function: ##EQU15## where: h is the number of the associated one of the hidden units, μ hi is the localized center of a h in the x 1 dimension, and f hi is a localized function in the x i dimension and non-localized in at least one other of the x i dimensions; interconnecting the output of select ones of the input vectors x i to the inputs of the hidden units in accordance with an interconnection scheme; providing a plurality of output units in an output layer, each of the output units for providing a single output y i for i ranging from 1 to M, M being an integer, each of the output units having associated therewith a predetermined control function g; and interconnecting the output of select ones of the hidden units to the inputs of predetermined ones of the M output units in accordance with the predetermined interconnection scheme, each of the interconnections between the output of the hidden units and the associated one of the M output units having an output weight associated therewith.
Claim 13
Dependent Claim
37 words
The method of claim 10 wherein the activation function of each of the hidden units is defined as a Gaussian function according to the following equation: ##EQU16## where σ hi =the standard deviation of a Gaussian function.
Claim 14
Dependent Claim
22 words
The method of claim 10 wherein the function f hi is non-localized for all of the dimensions other than the i dimension.
Claim 15
Dependent Claim
30 words
The method of claim 10 and further comprising weighting the interconnection between each of the input vectors x i and the interconnected one of the inputs of the hidden units.
Claim 16
Dependent Claim
18 words
The method of claim 10 wherein the function g in the M output units comprises a sigmoidal function.
Claim 17
Dependent Claim
15 words
The method of claim 10 and further comprising varying the value of the output weights.
Claim 18
Dependent Claim
96 words
The method of claim 15 and further comprising: inputting a plurality of known input values to the input vectors x i , each of the known input values having associated therewith a known target output value; comparing the known target value with the actual output value from the output units for a given one of the known input values input to the network; determining the error between the actual output value and the associated known target value; and varying the value of μ hi in the activation function and the output weights to minimize the error.
Claim 19
Dependent Claim
24 words
The method of claim 16 wherein the step of minimizing the error comprises minimizing the error in accordance with a Back Propagation learning algorithm.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1995-05-12 | 1995-11-11 | 1996-05-12 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1999-05-12 | 1999-11-11 | 2000-05-12 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2003-05-12 | 2003-11-11 | 2004-05-12 | $7700.00 | Window Closed / Lapsed |