D. P. Helmbold, S. Panizza, and M. K. Warmuth, "Direct and indirect algorithms for on-line learning of disjunctions," Theor. Comput. Sci., vol. 284, pp. 109--142, 2002. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Tracking a Small Set of Experts by Mixing Past Posteriors - Bousquet, Warmuth (2002) (2 citations) Self-citation (Warmuth) (Correct)
No context found.
D. Helmbold, S. Panizza, M. K. Warmuth. Direct and indirect algorithms for on-line learning of disjunctions. In Paul Fischer and Hans Ulrich Simon, editors, Computational Learning Theory: 4th European Conference (EuroCOLT '99), Berlin, March 1999. Springer.
Predicting Nearly as Well as the Best Pruning of a Planar.. - Takimoto, Warmuth (2000) (5 citations) Self-citation (Warmuth) (Correct)
....exponentially large. So the goal is to nd ecient implementations of the direct algorithms so that the exponentially many weights are implicitly maintained. In the case of trees this is possible [HS97,TMV98] Other applications that simulate algorithms with exponentially many weights are given in [HPW99,MW98] We now sketch how this can be done when the decision dag is planar. 3 Recall that each pruning and path intersect at one edge. Therefore in each trial the edges on the path determine the predictions of all the prunings as well as their losses. So in each trial the edges on the path also ....
D. Helmbold, S. Panizza, and M. Warmuth. Direct and indirect algorithm for on-line learning of disjunctions. In 4th EuroCOLT, pages 138-152, 1999.
Predicting Nearly as Well as the Best Pruning of a Planar.. - Takimoto, Warmuth (1999) (5 citations) Self-citation (Warmuth) (Correct)
....exponentially large. So the goal is to find efficient implementations of the direct algorithms so that the exponentially many weights are implicitly maintained. In the case of trees this is possible [HS97] Other applications that simulate algorithms with exponentially many weights are given in [HPW99, MW98] We now sketch how this can be done when the decision dag is planar. Recall that each pruning and path intersect at one edge. Therefore in each trial the edges on the path determine the predictions of all the prunings as well as their losses. So in each trial the edges on the path also ....
D. Helmbold, S. Panizza, and M. Warmuth. Direct and indirect algorithm for on-line learning of disjunctions. 4th EuroCOLT, 138--152, 1999.
An Incremental Learning Algorithm with Confidence.. - Polikar, Udpa, Udpa.. (2004) (Correct)
No context found.
D. P. Helmbold, S. Panizza, and M. K. Warmuth, "Direct and indirect algorithms for on-line learning of disjunctions," Theor. Comput. Sci., vol. 284, pp. 109--142, 2002.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC