Computer Science – Artificial Intelligence
Scientific paper
2008-06-26
LNCS 5010, pp. 302-313, 2008
Computer Science
Artificial Intelligence
Scientific paper
The remarkable results of Foster and Vohra was a starting point for a series of papers which show that any sequence of outcomes can be learned (with no prior knowledge) using some universal randomized forecasting algorithm and forecast-dependent checking rules. We show that for the class of all computationally efficient outcome-forecast-based checking rules, this property is violated. Moreover, we present a probabilistic algorithm generating with probability close to one a sequence with a subsequence which simultaneously miscalibrates all partially weakly computable randomized forecasting algorithms. %subsequences non-learnable by each randomized algorithm. According to the Dawid's prequential framework we consider partial recursive randomized algorithms.
No associations
LandOfFree
On Sequences with Non-Learnable Subsequences 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 On Sequences with Non-Learnable Subsequences, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and On Sequences with Non-Learnable Subsequences will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-246656