Online Non-Additive Path Learning under Full and Partial Information

Corinna Cortes, Vitaly Kuznetsov, Mehryar Mohri, Holakou Rahmanian, Manfred Warmuth
; Proceedings of the 30th International Conference on Algorithmic Learning Theory, PMLR 98:274-299, 2019.

Abstract

We study the problem of online path learning with non-additive gains, which is a central problem appearing in several applications, including ensemble structured prediction. We present new online algorithms for path learning with non-additive count-based gains for the three settings of full information, semi-bandit and full bandit with very favorable regret guarantees. A key component of our algorithms is the definition and computation of an intermediate context-dependent automaton that enables us to use existing algorithms designed for additive gains. We further apply our methods to the important application of ensemble structured prediction. Finally, beyond count-based gains, we give an efficient implementation of the EXP3 algorithm for the full bandit setting with an arbitrary (non-additive) gain.

Cite this Paper


BibTeX
@InProceedings{pmlr-v98-cortes19a, title = {Online Non-Additive Path Learning under Full and Partial Information}, author = {Cortes, Corinna and Kuznetsov, Vitaly and Mohri, Mehryar and Rahmanian, Holakou and Warmuth, Manfred}, booktitle = {Proceedings of the 30th International Conference on Algorithmic Learning Theory}, pages = {274--299}, year = {2019}, editor = {Aurélien Garivier and Satyen Kale}, volume = {98}, series = {Proceedings of Machine Learning Research}, address = {Chicago, Illinois}, month = {22--24 Mar}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v98/cortes19a/cortes19a.pdf}, url = {http://proceedings.mlr.press/v98/cortes19a.html}, abstract = { We study the problem of online path learning with non-additive gains, which is a central problem appearing in several applications, including ensemble structured prediction. We present new online algorithms for path learning with non-additive count-based gains for the three settings of full information, semi-bandit and full bandit with very favorable regret guarantees. A key component of our algorithms is the definition and computation of an intermediate context-dependent automaton that enables us to use existing algorithms designed for additive gains. We further apply our methods to the important application of ensemble structured prediction. Finally, beyond count-based gains, we give an efficient implementation of the EXP3 algorithm for the full bandit setting with an arbitrary (non-additive) gain.} }
Endnote
%0 Conference Paper %T Online Non-Additive Path Learning under Full and Partial Information %A Corinna Cortes %A Vitaly Kuznetsov %A Mehryar Mohri %A Holakou Rahmanian %A Manfred Warmuth %B Proceedings of the 30th International Conference on Algorithmic Learning Theory %C Proceedings of Machine Learning Research %D 2019 %E Aurélien Garivier %E Satyen Kale %F pmlr-v98-cortes19a %I PMLR %J Proceedings of Machine Learning Research %P 274--299 %U http://proceedings.mlr.press %V 98 %W PMLR %X We study the problem of online path learning with non-additive gains, which is a central problem appearing in several applications, including ensemble structured prediction. We present new online algorithms for path learning with non-additive count-based gains for the three settings of full information, semi-bandit and full bandit with very favorable regret guarantees. A key component of our algorithms is the definition and computation of an intermediate context-dependent automaton that enables us to use existing algorithms designed for additive gains. We further apply our methods to the important application of ensemble structured prediction. Finally, beyond count-based gains, we give an efficient implementation of the EXP3 algorithm for the full bandit setting with an arbitrary (non-additive) gain.
APA
Cortes, C., Kuznetsov, V., Mohri, M., Rahmanian, H. & Warmuth, M.. (2019). Online Non-Additive Path Learning under Full and Partial Information. Proceedings of the 30th International Conference on Algorithmic Learning Theory, in PMLR 98:274-299

Related Material