%Aigaion2 BibTeX export from Idiap Publications %Sunday 22 December 2024 04:17:55 AM @INPROCEEDINGS{grandvalet:nips:2005, author = {Grandvalet, Yves and Mari{\'{e}}thoz, Johnny and Bengio, Samy}, projects = {Idiap}, title = {A Probabilistic Interpretation of SVMs with an Application to Unbalanced Classification}, booktitle = {Advances in Neural Information Processing Systems, {NIPS} 15}, year = {2005}, note = {IDIAP-RR 05-26}, crossref = {grandvalet:rr05-26}, abstract = {In this paper, we show that the hinge loss can be interpreted as the neg-log-likelihood of a semi-parametric model of posterior probabilities. From this point of view, SVMs represent the parametric component of a semi-parametric model fitted by a maximum a posteriori estimation procedure. This connection enables to derive a mapping from SVM scores to estimated posterior probabilities. Unlike previous proposals, the suggested mapping is interval-valued, providing a set of posterior probabilities compatible with each SVM score. This framework offers a new way to adapt the SVM optimization problem when decisions result in unequal losses. Experiments on an unbalanced classification loss show improvements over state-of-the-art procedures.}, pdf = {https://publications.idiap.ch/attachments/papers/2005/grandvalet-nips-2005.pdf}, postscript = {ftp://ftp.idiap.ch/pub/papers/2005/grandvalet-nips-2005.ps.gz}, ipdmembership={learning}, } crossreferenced publications: @TECHREPORT{grandvalet:rr05-26, author = {Grandvalet, Yves and Mari{\'{e}}thoz, Johnny and Bengio, Samy}, projects = {Idiap}, title = {A Probabilistic Interpretation of SVMs with an Application to Unbalanced Classification}, type = {Idiap-RR}, number = {Idiap-RR-26-2005}, year = {2005}, institution = {IDIAP}, note = {Published in Advances in Neural Information Processing Systems, {NIPS} 15, 2005}, abstract = {In this paper, we show that the hinge loss can be interpreted as the neg-log-likelihood of a semi-parametric model of posterior probabilities. From this point of view, SVMs represent the parametric component of a semi-parametric model fitted by a maximum a posteriori estimation procedure. This connection enables to derive a mapping from SVM scores to estimated posterior probabilities. Unlike previous proposals, the suggested mapping is interval-valued, providing a set of posterior probabilities compatible with each SVM score. This framework offers a new way to adapt the SVM optimization problem when decisions result in unequal losses. Experiments on an unbalanced classification loss show improvements over state-of-the-art procedures.}, pdf = {https://publications.idiap.ch/attachments/reports/2005/grandvalet-idiap-rr-05-26.pdf}, postscript = {ftp://ftp.idiap.ch/pub/reports/2005/grandvalet-idiap-rr-05-26.ps.gz}, ipdmembership={learning}, }