A Theory for Neural Networks with Time Delays

By Bert J. M. de Vries, José Carlos Príncipe. 1990

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

Deep learning in neural networks: An overview (2014, neutral): “Ideas for dealing with long time lags and deep CAPs To deal with long time lags between relevant events, sev- eral sequence processing methods were proposed, including Focused BP based on decay factors for activations of units in RNNs (Mozer, 1989, 1992), Time-Delay Neural Networks (TDNNs) (Lang, Wa”
Gradient Flow in Recurrent Nets: the Difficulty of Learning Long-Term Dependencies (2001, methodology): “To deal with long time lags, Mozer [18] uses time constants influencing changes of unit activations (deVries and Principe’s related approach [8] may be viewed as a mixture of time-delay neural networks (TDNN) [15] and time constants).”
Monthly catch forecasting of anchovy Engraulis ringens in the north area of Chile: Non-linear univariate approach (2007, methodology): “In order to verify if more accurate CNN solutions could be achieved, three variations of the typical autoregressive univariate CNN models or additions to those models were tested, namely: (a) a convolution process of the input variables to the CNN model ( De Vries and Principe, 1991 ); (b) recurrent”
Neural net architectures for temporal sequence processing (2007, background): “de Vries and Principe (1991) show that the gamma kernels form a basis set, meaning that for an arbitrary kernel function c that decays exponentially to 0 as τ →∞, there exists a value of Ω and a set of coefficients {qj} such that Equation 1 holds.”
Bridging Long Time Lags by Weight Guessing and \Long Short Term Memory" (1996, background): “…tasks, though, are more di(cid:14)cult than the above benchmark problems: they cannot be solved at all in reasonable time by random search (we tried various architectures) nor any other recurrent net learning algorithm we are aware of, e.g., [25, 20, 7, 8, 33, 14, 21, 5, 15, 29, 1, 24, 26, 19].”
Long Short-Term Memory (1997, background): “weighted sum of old activations (see also de Vries & Principe, 1991 ).”
Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Network (2018, background): “, to account for delayed neural interactions) [10].”