&lt;title&gt;Adaptive time-delay neural network for temporal correlation and prediction&lt;/title&gt;

Lin, Daw-Tung; Dayhoff, Judith E.; Ligomenides, Panos A.

doi:10.1117/12.131597

Cited by 30 publications

(8 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a special case of the so-called Hahn polynomials, the discrete Chebyshev polynomials fulfill the three-term recurrence relation given in (5) and (6) with for (8) for (9) For a degree up to five, these polynomials are set out in Table I. Their squared norms are given by (10) for . The approximation problem "find a polynomial of degree which minimizes " (11) can now be solved using an orthogonal expansion by (12) Note that the approximation problem is defined over discrete (and equidistant) points in time, whereas its solution is defined (and can be evaluated) over .…”

Section: A Least Squares Approximation Of a Time Series With Orthogomentioning

confidence: 99%

See 1 more Smart Citation

Processing Short-Term and Long-Term Information With a Combination of Polynomial Approximation Techniques and Time-Delay Neural Networks

Fuchs

Gruber²,

Reitmaier

et al. 2009

IEEE Trans. Neural Netw.

View full text Add to dashboard Cite

Neural networks are often used to process temporal information, i.e., any kind of information related to time series. In many cases, time series contain short-term and long-term trends or behavior. This paper presents a new approach to capture temporal information with various reference periods simultaneously. A least squares approximation of the time series with orthogonal polynomials will be used to describe short-term trends contained in a signal (average, increase, curvature, etc.). Long-term behavior will be modeled with the tapped delay lines of a time-delay neural network (TDNN). This network takes the coefficients of the orthogonal expansion of the approximating polynomial as inputs such considering short-term and long-term information efficiently. The advantages of the method will be demonstrated by means of artificial data and two real-world application examples, the prediction of the user number in a computer network and online tool wear classification in turning.

show abstract

Section: A Least Squares Approximation Of a Time Series With Orthogomentioning

confidence: 99%

“…Paradigms applying the former mechanism are, for example, TDNNs (also called FIR networks) [6], [7], adaptive time-delay neural networks (ATNNs) [10], or radial basis function networks with time-delays (TDRBFs) [11]. These networks make use of local temporal information in a limited time interval to compute an output.…”

Section: Related Workmentioning

confidence: 99%

Processing Short-Term and Long-Term Information With a Combination of Polynomial Approximation Techniques and Time-Delay Neural Networks

Fuchs

Gruber²,

Reitmaier

et al. 2009

IEEE Trans. Neural Netw.

View full text Add to dashboard Cite

show abstract

“…Paradigms applying the first mechanism are TDNN (also called FIR-networks) [196,198], adaptive time-delay neural networks (ATNN) [120], or radial basis function networks with time-delays (TDRBF) [86], for instance. These networks make use of local temporal information in a limited time interval to compute an output, i.e.…”

Section: Temporal Informationmentioning

confidence: 99%

On-Line and Indirect Tool Wear Monitoring in Turning With Artificial Neural Networks: A Review of More Than a Decade of Research

Sick

2002

Mechanical Systems and Signal Processing

371

143

View full text Add to dashboard Cite

“…An Adaptive Time Delay Neural Network model has been proposed (ATNN) [14,6] to overcome this limitation and to process temporally modulated signals. This network adapts its time delay values as well as its weights during training, to better accommodate to changing temporal patterns, and to provide more flexibility for optimization tasks.…”

Section: Introductionmentioning

confidence: 99%

<title>Spatiotemporal topology and temporal sequence identification with an adaptive time-delay neural network</title>

1993

Self Cite

View full text Add to dashboard Cite

Inspired from the time delays that occur in neurobiological signal transmission, we describe an Adaptive Time Delay Neural Network (ATNN) which is a powerful dynamic learning technique for spatiotemporal pattern transformation and temporal sequence identification. The dynamic properties of this network are formulated through the adaptation of time-delays and synapse weights, which are adjusted on-line based on gradient descent rules according to the evolution of observed inputs and outputs. We have applied the ATNN to examples that possess spatiotemporal complexity, with temporal sequences that are completed by the network. The ATNN is able to applied to pattern completion. Simulation results show that the ATNN learns the topology of a circular and figure eight trajectories within 500 on-line training iterations, and reproduces the trajectory dynamically with very high accuracy. The ATNN was also trained to model the Fourier series expansion of the sum of different odd harmonics. The resulting network provides more flexibility and efficiency than the TDNN and allows the network to seek optimal values for time-delays as well as optimal synapse weights.

show abstract

<title>Adaptive time-delay neural network for temporal correlation and prediction</title>

Cited by 30 publications

References 0 publications

Processing Short-Term and Long-Term Information With a Combination of Polynomial Approximation Techniques and Time-Delay Neural Networks

Processing Short-Term and Long-Term Information With a Combination of Polynomial Approximation Techniques and Time-Delay Neural Networks

On-Line and Indirect Tool Wear Monitoring in Turning With Artificial Neural Networks: A Review of More Than a Decade of Research

<title>Spatiotemporal topology and temporal sequence identification with an adaptive time-delay neural network</title>

Contact Info

Product

Resources

About