Leading strategies in competitive on-line prediction

Details Leading strategies in competitive on-line prediction Leading strategies in competitive on-line prediction

2006-07-27

arxiv.org/abs/cs/0607134v1

Computer Science

Learning

20 pages; a conference version is to appear in the ALT'2006 proceedings

Scientific paper

We start from a simple asymptotic result for the problem of on-line regression with the quadratic loss function: the class of continuous limited-memory prediction strategies admits a "leading prediction strategy", which not only asymptotically performs at least as well as any continuous limited-memory strategy but also satisfies the property that the excess loss of any continuous limited-memory strategy is determined by how closely it imitates the leading strategy. More specifically, for any class of prediction strategies constituting a reproducing kernel Hilbert space we construct a leading strategy, in the sense that the loss of any prediction strategy whose norm is not too large is determined by how closely it imitates the leading strategy. This result is extended to the loss functions given by Bregman divergences and by strictly proper scoring rules.

Affiliated with

Also associated with

No associations

Rating

Leading strategies in competitive on-line prediction 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 Leading strategies in competitive on-line prediction, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Leading strategies in competitive on-line prediction will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-250572