A decision-theoretic generalization of on-line learning and an application to boosting. In the first part of the paper we consider the problem of dynamically apportioning resources among a set of options in a worst-case on-line framework. The model we study can be interpreted as a broad, abstract extension of the well-studied on-line prediction model to a general decision-theoretic setting. We show that the multiplicative weight-update Littlestone-Warmuth rule can be adapted to this model, yielding bounds that are slightly weaker in some cases, but applicable to a considerably more general class of learning problems. We show how the resulting learning algorithm can be applied to a variety of problems, including gambling, multiple-outcome prediction, repeated games, and prediction of points in $bfR^n$. In the second part of the paper we apply the multiplicative weight-update technique to derive a new boosting algorithm. This boosting algorithm does not require any prior knowledge about the performance of the weak learning algorithm. We also study generalizations of the new boosting algorithm to the problem of learning functions whose range, rather than being binary, is an arbitrary finite set or a bounded segment of the real line.

References in zbMATH (referenced in 364 articles , 1 standard article )

Showing results 1 to 20 of 364.
Sorted by year (citations)

1 2 3 ... 17 18 19 next

  1. Darnstädt, Malte; Ries, Christoph; Simon, Hans Ulrich: Hierarchical design of fast minimum disagreement algorithms (2018)
  2. Mukhopadhyay, Subhadeep; Nandi, Shinjini: LPiTrack: eye movement pattern recognition algorithm and application to biometric identification (2018)
  3. Orabona, Francesco; Pál, Dávid: Scale-free online learning (2018)
  4. Wang, Boxiang; Zou, Hui: Another look at distance-weighted discrimination (2018)
  5. Wang, Zhu: Robust boosting with truncated loss functions (2018)
  6. Alon, Noga; Cesa-Bianchi, Nicolò; Gentile, Claudio; Mannor, Shie; Mansour, Yishay; Shamir, Ohad: Nonstochastic multi-armed bandits with graph-structured feedback (2017)
  7. Denis, Christophe; Hebiri, Mohamed: Confidence sets with expected sizes for multiclass classification (2017)
  8. Edhan, Omer; Hellman, Ziv; Sherill-Rofe, Dana: Sex with no regrets: how sexual reproduction uses a no regret learning algorithm for evolutionary advantage (2017)
  9. Feige, Uriel; Koren, Tomer; Tennenholtz, Moshe: Chasing ghosts: competing with stateful policies (2017)
  10. Gotoh, Jun-ya; Uryasev, Stan: Support vector machines based on convex risk functions and general norms (2017)
  11. Gutfreund, Dan; Kontorovich, Aryeh; Levy, Ran; Rosen-Zvi, Michal: Boosting conditional probability estimators (2017)
  12. Hua, Jia-Chen; Noorian, Farzad; Moss, Duncan; Leong, Philip H. W.; Gunaratne, Gemunu H.: High-dimensional time series prediction using kernel-based koopman mode regression (2017)
  13. Krauss, Christopher; Do, Xuan Anh; Huck, Nicolas: Deep neural networks, gradient-boosted trees, random forests: statistical arbitrage on the S&P 500 (2017)
  14. Mcmahan, H. Brendan: A survey of algorithms and analysis for adaptive online learning (2017)
  15. Proença, Hugo; Neves, João C.: Fusing vantage point trees and linear discriminants for fast feature classification (2017)
  16. Santos, Eugene jun.; Zhao, Yan: Automatic emergence detection in complex systems (2017)
  17. Siciliano, Roberta; D’Ambrosio, Antonio; Aria, Massimo; Amodio, Sonia: Analysis of web visit histories. II: Predicting navigation by nested STUMP regression trees (2017)
  18. Srinivasan, Ashwin; Bain, Michael: An empirical study of on-line models for relational data streams (2017)
  19. Adamskiy, Dmitry; Koolen, Wouter M.; Chernov, Alexey; Vovk, Vladimir: A closer look at adaptive regret (2016)
  20. Ali, Farman; Hayat, Maqsood: Machine learning approaches for discrimination of extracellular matrix proteins using hybrid feature space (2016)

1 2 3 ... 17 18 19 next