Reverse time delay neural network for pattern generation

Details Reverse time delay neural network for pattern generation Reverse time delay neural network for pattern generation

1992-06-30

1995-05-02

MacDonald, Allen R.

395 62, 395 22, 395 23, G06F 1518

Patent

active

054127548

ABSTRACT:
Trajectories are generated in response to an input label by using a reverse time delay neural network. The reverse time delay neural network comprises an input layer, a plurality of hidden layers, and an output layer, all arranged in succession so that the number of frames per layer increases as the network is traversed from the input layer to the output layer. Additionally, the number of features decreases as the network is traversed from the input layer to the output layer. Features of the trajectory are created from the input label so that a time series of frames can be output by the network. Frames generally relate to particular epochs of time or time units and a frame includes a plurality of features.
Interconnection between layers is accomplished using differential neuron units. In the differential neuron unit, standard neuron weighting, summing, and nonlinear squashing functions are performed on the inputs thereto. Moreover, the output of the differential neuron unit includes a contribution from the value of the corresponding feature in the previous frame in the same layer as the differential neuron unit output.

REFERENCES:
patent: 4752906 (1988-06-01), Kleinfeld
patent: 4852010 (1989-07-01), Grossberg et al.
patent: 4874963 (1989-10-01), Alspector
patent: 4990838 (1991-02-01), Kawato et al.
patent: 5105468 (1992-04-01), Guyon et al.
Parallelism, Hierarchi, Scaling in Time-Delay NN for Spotting Japanese Phonemes/CV-Syllables 18-22/Jun./1989, H Sawai et al. pp. II-81-88.
Adaptation and Learning for Robotic Manipulator by NN T. Fukuda et al. 5-7 Dec. 1990 pp. 3283-3288.
{0,1}.sup.n Space Self-Organising Feature Maps--Extensions and Hardware Implementations Allinson et al. IEE Conf. Publ. 1989 Oct. 16-18.
M. I. Jordan, Computer and Information Science, Univ. of Mass. at Amherst, May 1988, "Supervised Learning and Systems with Excess Degrees of Freedom," pp. 13-17.
B. A. Pearlmutter, Neural Computation, vol. 1, 1989, "Learning State Space Trajectories in Recurrent Neural Networks," pp. 263-269.
A. Waibel et al., IEEE Trans. on Acoustics, Speech, and Signal Proc., vol. 37, No. 3, Mar. 1989, "Phoneme Recognition using Time-Delay Neural Networks," pp. 328-339.
C. C. Tappert et al., IEEE Trans. on Pattern Analysis and Machine Intell., vol. 12, No. 8, Aug. 1990, "The State of the Art in On-Line Handwriting Recognition," pp 787-808.
I. Guyon et al., Pattern Recognition, vol. 24, No. 2, 1991 "Design of a Neural Network Character Recognizer for a Touch Terminal," pp. 105-119.

Affiliated with

Also associated with

Rating

Reverse time delay neural network for pattern generation does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Reverse time delay neural network for pattern generation, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Reverse time delay neural network for pattern generation will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-1144358