ejournals

Information

Afilliation

University of Bielsko-Biala Department of Computer Science

Articles

Sort:

Author

Data Selection for Neural Networks

Mirosław Kordos

Schedae Informaticae, Volume 25, 2016, pp. 153 - 164

https://doi.org/10.4467/20838476SI.16.012.6193

Several approaches to joined feature and instance selection in neural network leaning are discussed and experimentally evaluated in respect to classification accuracy and dataset compression, considering also their computational complexity. These include various versions of feature and instance selection prior to the network learning, the selection embedded in the neural network and hybrid approaches, including solutions developed by us. The advantages and disadvantages of each approach are discussed and some possible improvements are proposed.

PDF

Author

Effectiveness of Unsupervised Training in Deep Learning Neural Networks

Mirosław Kordos

Schedae Informaticae, Volume 24, 2015, pp. 41 - 51

https://doi.org/10.4467/20838476SI.15.004.3026

Deep learning is a field of research attracting nowadays much attention, mainly because deep architectures help in obtaining outstanding results on many vision, speech and natural language processing – related tasks. To make deep learning effective, very often an unsupervised pretraining phase is applied. In this article, we present experimental study evaluating usefulness of such approach, testing on several benchmarks and different percentages of labeled data, how Contrastive Divergence (CD), one of the most popular pretraining methods, influences network generalization.

PDF