Time series prediction by using a connectionist network with internal delay lines

By Eric A. Wan. 1993

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Time-series forecasting with deep learning: a survey (2020, methodology): “Furthermore, older models of neural networks have been used historically in time-series applications, as seen in [24] and [25].”
Deep learning in neural networks: An overview (2014, background): “Notably, an NN with internal delay lines won the Santa Fe time-series competition on chaotic intensity pulsations of an NH3 laser (Wan, 1994; Weigend and Gershenfeld, 1993).”
Dynamic Factor Graphs for Time Series Modeling (2009, methodology): “Typically, one can use simple multivariate autoregressive linear functions to map the state variables, or can also resort to nonlinear dynamics modeled by a Convolutional Network [8] with convolutions (FIR ﬁlters) across time, as in Time-Delay Neural Networks [7,18].”
Recurrent Neural Networks for Prediction: Learning Algorithms, Architectures and Stability (2001, background): “Takens’ theorem (Wan 1993) implies that for a wide class of deterministic systems, there exists a diffeomorphism (one-to-one differential mapping) between a finite window of the time series and the underlying”
Learning long-term dependencies in NARX recurrent neural networks (1996, result): “It may also be possible to obtain similar results for the architectures proposed in [6, 9, 24, 37].”
Artificial neural network modeling of water table depth fluctuations (2001, background): “A particular feature of the IDNN is that the memory structure is focused on the input layer; this makes it different from the general time delay neural network (TDNN) [ Waibel et al ., ; Wan , 1994], which uses internal delays at each neuron.”
Granular neural web agents for stock prediction (2002, background): “The number of necessary weights can sometimes be significantly reduced by assuming an implicit symmetry in the optimal network structure [15].”