ML Anthology

Aggregation by Exponential Weighting and Sharp Oracle Inequalities

Arnak S. Dalalyan, Alexandre B. Tsybakov
COLT 2007 pp. 97-111
doi:10.1007/978-3-540-72927-3_9 /colt/2007/dalalyan2007colt-aggregation/

Abstract

In the present paper, we study the problem of aggregation under the squared loss in the model of regression with deterministic design. We obtain sharp oracle inequalities for convex aggregates defined via exponential weights, under general assumptions on the distribution of errors and on the functions to aggregate. We show how these results can be applied to derive a sparsity oracle inequality.

PDF COLT Semantic Scholar

Cite

Text

Dalalyan and Tsybakov. "Aggregation by Exponential Weighting and Sharp Oracle Inequalities." Annual Conference on Computational Learning Theory, 2007. doi:10.1007/978-3-540-72927-3_9

Markdown

[Dalalyan and Tsybakov. "Aggregation by Exponential Weighting and Sharp Oracle Inequalities." Annual Conference on Computational Learning Theory, 2007.](https://mlanthology.org/colt/2007/dalalyan2007colt-aggregation/) doi:10.1007/978-3-540-72927-3_9

BibTeX

@inproceedings{dalalyan2007colt-aggregation,
  title     = {{Aggregation by Exponential Weighting and Sharp Oracle Inequalities}},
  author    = {Dalalyan, Arnak S. and Tsybakov, Alexandre B.},
  booktitle = {Annual Conference on Computational Learning Theory},
  year      = {2007},
  pages     = {97-111},
  doi       = {10.1007/978-3-540-72927-3_9},
  url       = {https://mlanthology.org/colt/2007/dalalyan2007colt-aggregation/}
}