Anticipation-based temporal pattern generation

Wang, D.L.; Yuwono, Bambang Darmo

doi:10.1109/21.370192

Cited by 37 publications

(23 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, Wang and Yuwono (1995) proposed the ABM model to improve the dual neuron model. By employing competitive learning, the selection of context detectors in the ABM model is self-organized.…”

Section: Extending To Dnn2mentioning

confidence: 99%

See 1 more Smart Citation

Learning temporal sequences using dual‐weight neurons

Wang

Tsai²,

2001

Journal of the Chinese Institute of Engineers

View full text Add to dashboard Cite

This paper considers the use of neural networks (NN's) in learning temporal sequence recognition and reproduction for which the sequence degree is unknown. This approach uses the output ambiguity to train the network without the need to assume or construct a separate model for the input sequence degree. First we introduce a primitive network called the DNN, comprising a plurality of dual-weight (DN) neurons. Each neuron is linked to other neurons by a long-term excitatory weight and a short-term inhibitory weight. Fast learning is made possible by employing a two-pass training rule to encode the temporal distance between two arbitrary pattern occurrences. The resulting DNN is then extended into a more generalized model, namely the DNN2. By incorporating the two-pass rule and a self-organizing algorithm, the DNN2 can achieve autonomous temporal sequence recognition and reproduction. Using training efficiency and hardware complexity criteria, the DNN networks are also contrasted with the work of Wang and Yuwono (1995).

show abstract

Section: Extending To Dnn2mentioning

confidence: 99%

“…By incorporating the two-pass rule and a self-organizing algorithm, the DNN2 can achieve autonomous temporal sequence recognition and reproduction. Using training efficiency and hardware complexity criteria, the DNN networks are also contrasted with the work of Wang and Yuwono (1995).…”

mentioning

confidence: 99%

Learning temporal sequences using dual‐weight neurons

Wang

Tsai²,

2001

Journal of the Chinese Institute of Engineers

View full text Add to dashboard Cite

show abstract

“…Since we loop over all elements in waitsðE i Þ, it is inappropriate to add to it from within the loop when ½j þ 1 ¼ E i . Hence, in such cases, ð; j þ 1Þ is added to a temporary list, called bag, which is later emptied into waitsðE i Þ on coming out of the loop (lines [12][13]21).…”

Section: Counting Nonoverlapping Occurrencesmentioning

confidence: 99%

“…It is clear by inspection that the second expression is maximized at ¼ 1 and the maximum value is ð 1 M Þ T . If P ðo; q Ã j ÃÞ given by (12) is maximized at ¼ 1, then the best EGH association for would be one with ¼ 1, which is merely a random iid model. We discuss this case later on.…”

Section: Mþ1mentioning

confidence: 99%

See 1 more Smart Citation

Discovering frequent episodes and learning hidden Markov models: a formal connection

Laxman

Sastry

Unnikrishnan

2005

IEEE Trans. Knowl. Data Eng.

101

View full text Add to dashboard Cite

Abstract-This paper establishes a formal connection between two common, but previously unconnected methods for analyzing data streams: discovering frequent episodes in a computer science framework and learning generative models in a statistics framework. We introduce a special class of discrete Hidden Markov Models (HMMs), called Episode Generating HMMs (EGHs), and associate each episode with a unique EGH. We prove that, given any two episodes, the EGH that is more likely to generate a given data sequence is the one associated with the more frequent episode. To be able to establish such a relationship, we define a new measure of frequency of an episode, based on what we call nonoverlapping occurrences of the episode in the data. An efficient algorithm is proposed for counting the frequencies for a set of episodes. Through extensive simulations, we show that our algorithm is both effective and more efficient than current methods for frequent episode discovery. We also show how the association between frequent episodes and EGHs can be exploited to assess the significance of frequent episodes discovered and illustrate empirically how this idea may be used to improve the efficiency of the frequent episode discovery.

show abstract

Sequence Learning

2011

Self‐Adaptive Systems for Machine Intelligence

View full text Add to dashboard Cite

Anticipation-based temporal pattern generation

Cited by 37 publications

References 33 publications

Learning temporal sequences using dual‐weight neurons

Learning temporal sequences using dual‐weight neurons

Discovering frequent episodes and learning hidden Markov models: a formal connection

Sequence Learning

Contact Info

Product

Resources

About