"Ignorance-based" systems

Gevins, A.; Morgan, Nelson

doi:10.1109/icassp.1984.1172798

Cited by 8 publications

(2 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The algorithm can thus concentrate on only promising branches and find the optimal feature subset, given sufficient time. In many real-life problems, however, reducing the number of features may actually reduce the classifier's error rate; thus, these problems are not monotonic [5,8]. This restriction and the complexity of the branch-and-bound procedure with high input dimensionality again have limited its application.…”

Section: Traditional Heuristically Guided Search Approachesmentioning

confidence: 99%

See 1 more Smart Citation

Using genetic algorithms to select and create features for pattern classification

Chang

Lippmann

Tong³

1990

1990 IJCNN International Joint Conference on Neural Networks

View full text Add to dashboard Cite

Section: Traditional Heuristically Guided Search Approachesmentioning

confidence: 99%

“…This may not be true with more complex problems that have nonmonotonic improvement in performance as more features are added [8,19] . The next problem with 153 input feature dimensions provided a more difficult problem domain.…”

Section: The Parallel Vector Problemmentioning

confidence: 99%

Using genetic algorithms to select and create features for pattern classification

Chang

Lippmann

Tong³

1990

1990 IJCNN International Joint Conference on Neural Networks

View full text Add to dashboard Cite

Continuous speech recognition by connectionist statistical methods

Bourlard

Morgan

1993

IEEE Trans. Neural Netw.

View full text Add to dashboard Cite

Over the period of 1987-1991, a series of theoretical and experimental results have suggested that multilayer perceptrons (MLP) are an effective family of algorithms for the smooth estimation of high-dimension probability density functions that are useful in continuous speech recognition. The early form of this work has focused on hidden Markov models (HMM) that are independent of phonetic context. More recently, the theory has been extended to context-dependent models. The authors review the basic principles of their hybrid HMM/MLP approach and describe a series of improvements that are analogous to the system modifications instituted for the leading conventional HMM systems over the last few years. Some of these methods directly trade off computational complexity for reduced requirements of memory and memory bandwidth. Results are presented on the widely used Resource Management speech database that has been distributed by the US National Institute of Standards and Technology.

show abstract

Speech recognition using weighted HMM and subspace projection approaches

Lee

1994

IEEE Trans. Speech Audio Process.

View full text Add to dashboard Cite

"Ignorance-based" systems

Cited by 8 publications

References 20 publications

Using genetic algorithms to select and create features for pattern classification

Using genetic algorithms to select and create features for pattern classification

Continuous speech recognition by connectionist statistical methods

Speech recognition using weighted HMM and subspace projection approaches

Contact Info

Product

Resources

About