Computer Science – Learning
Scientific paper
2005-07-18
Proc. 16th International Conf. on Algorithmic Learning Theory (ALT 2005) 356-370
Computer Science
Learning
15 LaTeX pages
Scientific paper
This paper shows how universal learning can be achieved with expert advice. To this aim, we specify an experts algorithm with the following characteristics: (a) it uses only feedback from the actions actually chosen (bandit setup), (b) it can be applied with countably infinite expert classes, and (c) it copes with losses that may grow in time appropriately slowly. We prove loss bounds against an adaptive adversary. From this, we obtain a master algorithm for "reactive" experts problems, which means that the master's actions may influence the behavior of the adversary. Our algorithm can significantly outperform standard experts algorithms on such problems. Finally, we combine it with a universal expert class. The resulting universal learner performs -- in a certain sense -- almost as well as any computable strategy, for any online decision problem. We also specify the (worst-case) convergence speed, which is very slow.
Hutter Marcus
Poland Jan
No associations
LandOfFree
Defensive Universal Learning with Experts does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Defensive Universal Learning with Experts, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Defensive Universal Learning with Experts will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-187963