By Gabriel B. P. Costa, Moacir Ponti, Alejandro C. Frery (auth.), Zhi-Hua Zhou, Friedhelm Schwenker (eds.)

This booklet constitutes the completely refereed revised chosen papers from the second one IAPR foreign Workshop, PSL 2013, held in Nanjing, China, in might 2013. the ten papers integrated during this quantity have been rigorously reviewed and chosen from 26 submissions. in part supervised studying is a swiftly evolving sector of computer studying. It generalizes many varieties of studying paradigms together with supervised and unsupervised studying, semi-supervised studying for class and regression, transductive studying, semi-supervised clustering, multi-instance studying, vulnerable label studying, coverage studying in in part observable environments, etc.

Show description

Read or Download Partially Supervised Learning: Second IAPR International Workshop, PSL 2013, Nanjing, China, May 13-14, 2013, Revised Selected Papers PDF

Similar international books

The Routledge International Handbook of Globalization Studies (Routledge International Handbooks)

The Routledge International Handbook of Globalization experiences deals scholars transparent and proficient chapters at the background of globalization and key theories that experience thought of the factors and effects of the globalization technique. There are noticeable sections demographic, financial, technological, social and cultural adjustments in globalization.

The Chern Symposium 1979: Proceedings of the International Symposium on Differential Geometry in honor of S.-S. Chern, held in Berkeley, California, June 1979

This quantity attests to the energy of differential geometry because it probes deeper into its inner constitution and explores ever widening connections with different topics in arithmetic and physics. To so much folks Professor S. S. Chern is sleek differential geometry, and we, his scholars, are thankful to him for best us to this fertile panorama.

Partially Supervised Learning: Second IAPR International Workshop, PSL 2013, Nanjing, China, May 13-14, 2013, Revised Selected Papers

This publication constitutes the completely refereed revised chosen papers from the second one IAPR overseas Workshop, PSL 2013, held in Nanjing, China, in could 2013. the ten papers incorporated during this quantity have been conscientiously reviewed and chosen from 26 submissions. partly supervised studying is a quickly evolving quarter of computer studying.

Extra info for Partially Supervised Learning: Second IAPR International Workshop, PSL 2013, Nanjing, China, May 13-14, 2013, Revised Selected Papers

Example text

Combining Pattern Classifiers: Methods and Algorithms. WileyInterscience, New York (2004) 18. : Speech emotion recognition using an enhanced co-training algorithm. In: IEEE International Conference on Multimedia and Expo, pp. 999–1002 (2007) 19. : The extended cohn-kanade dataset (ck+): a complete dataset for action unit and emotion-specified expression. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW 2010), pp. 94–101 (2010) 20. : Dictionary learning for the sparse modelling of atrial fibrillation in ecg signals.

In many real applications, the original data matrix only contains non-negative values such as the term frequency in document sets, the gray value of the image pixel and etc. In this paper, thus, we introduce non-negative matrix factorization (NMF) technique to model the loss function between the data samples and the given limited label information. The basic idea of NMF is to find several non-negative bases from the original data and then linearly project data onto these bases [7]. Meanwhile, the nonnegative constraint leads NMF to a parts-based representation of the sample, which makes NMF popular in many fields.

G. with different scaling) while dissimilar signals are represented by combinations of different dictionary atoms. For a classifier, this is a favorable starting position. For future work, the applicability of this approach to non-periodical signals could be examined with possible modifications to the alignment procedure. g. respiration) or even a single dictionary from combined signals. 6 Conclusion This paper presented an approach for emotion recognition based on blood volume pulse signals using sparse representations.

Download PDF sample

Rated 4.20 of 5 – based on 16 votes