Back error propagation - A critique

Juřík, Martin

doi:10.1109/cmpcon.1988.4895

Cited by 2 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Instead of repeating the local search from many different initial conditions, or applying perturbations on the weights when the search seems to stagnate, it is desirable to rely on a more robust, global search method. In 1988, Jurick provided a critique of BP and was among the firsts to suggest the use of robust tuning techniques for NN's, including evolutionary approaches [79].…”

Section: Nn's Generated By Ecmentioning

confidence: 99%

Hybrid soft computing systems: industrial and commercial applications

et al. 1999

View full text Add to dashboard Cite

Soft computing (SC) is an association of computing methodologies that includes as its principal members fuzzy logic, neurocomputing, evolutionary computing and probabilistic computing. We present a collection of methods and tools that can be used to perform diagnostics, estimation, and control. These tools are a great match for real-world applications that are characterized by imprecise, uncertain data and incomplete domain knowledge. We outline the advantages of applying SC techniques and in particular the synergy derived from the use of hybrid SC systems. We illustrate some combinations of hybrid SC systems, such as fuzzy logic controllers (FLC's) tuned by neural networks (NN's) and evolutionary computing (EC), NN's tuned by EC or FLC's, and EC controlled by FLC's. We discuss three successful real-world examples of SC applications to industrial equipment diagnostics, freight train control, and residential property valuation.

show abstract

Section: Nn's Generated By Ecmentioning

confidence: 99%

Hybrid soft computing systems: industrial and commercial applications

et al. 1999

View full text Add to dashboard Cite

show abstract

“…Kulkarni, Yap, and Bayers [6] used moments as input to neural networks for analyzing aircraft images. We used backpropagation [11,13] algorithm to train a multilayered neural network.…”

Section: Introduccrlonmentioning

confidence: 99%

Transformations and distortions tolerant recognition of numerals using neural networks

Srivastava

1991

Proceedings of the 19th Annual Conference on Computer Science - CSC '91

View full text Add to dashboard Cite

This paper presents a multilayered artificial neural network model to recognize numerals independent of their sizes, positions, and orientations. We used a modified backpropagation algorithm to train the network.We preprocessed the numerals for "feature" extraction by calculating their moments. The moments are invariant under translation, scaling, and rotation transformations. We used the moments as input to the network rather than the digitized pattern itself. The network was able to recognize transformed and slightly deformed numerals.Parallel computational capability of the network makes it an attractive alternative for real-time commercial applications.

show abstract

Back error propagation - A critique

Cited by 2 publications

References 6 publications

Hybrid soft computing systems: industrial and commercial applications

Hybrid soft computing systems: industrial and commercial applications

Transformations and distortions tolerant recognition of numerals using neural networks

Contact Info

Product

Resources

About