caa a22 4500 60547852X CHVBK 20210128100406.0 cr unu---uuuuu 210128e20150501xx s 000 0 eng 10.1007/s10994-015-5490-3 doi (NATIONALLICENCE)springer-10.1007/s10994-015-5490-3 Unconfused ultraconservative multiclass algorithms [Elektronische Daten] [Ugo Louche, Liva Ralaivola] We tackle the problem of learning linear classifiers from noisy datasets in a multiclass setting. The two-class version of this problem was studied a few years ago where the proposed approaches to combat the noise revolve around a Perceptron learning scheme fed with peculiar examples computed through a weighted average of points from the noisy training set. We propose to build upon these approaches and we introduce a new algorithm called Unconfused Multiclass additive Algorithm (U MA) which may be seen as a generalization to the multiclass setting of the previous approaches. In order to characterize the noise we use the confusion matrix as a multiclass extension of the classification noise studied in the aforementioned literature. Theoretically well-founded, U MA furthermore displays very good empirical noise robustness, as evidenced by numerical simulations conducted on both synthetic and real data. The Author(s), 2015 Multiclass classification nationallicence Perceptron nationallicence Noisy labels nationallicence Confusion Matrix nationallicence Ultraconservative algorithms nationallicence Louche Ugo Qarma, Lab. d'Informatique Fondamentale de Marseille, CNRS, Université d'Aix-Marseille, Marseille, France aut Ralaivola Liva Qarma, Lab. d'Informatique Fondamentale de Marseille, CNRS, Université d'Aix-Marseille, Marseille, France aut Machine Learning Springer US; http://www.springer-ny.com 99/2(2015-05-01), 327-351 0885-6125 99:2<327 2015 99 10994 https://doi.org/10.1007/s10994-015-5490-3 text/html Onlinezugriff via DOI BK010053 XK010053 XK010000 Metadata rights reserved Springer special CC-BY-NC licence nationallicence 1 research-article jats NATIONALLICENCE NATIONALLICENCE NL-springer NATIONALLICENCE 856 40 https://doi.org/10.1007/s10994-015-5490-3 text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Louche Ugo Qarma, Lab. d'Informatique Fondamentale de Marseille, CNRS, Université d'Aix-Marseille, Marseille, France aut NATIONALLICENCE 700 1- Ralaivola Liva Qarma, Lab. d'Informatique Fondamentale de Marseille, CNRS, Université d'Aix-Marseille, Marseille, France aut NATIONALLICENCE 773 0- Machine Learning Springer US; http://www.springer-ny.com 99/2(2015-05-01), 327-351 0885-6125 99:2<327 2015 99 10994