Optimal Online Prediction in Adversarial Environments

  • Peter L. Bartlett
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6331)


In many prediction problems, including those that arise in computer security and computational finance, the process generating the data is best modelled as an adversary with whom the predictor competes. Even decision problems that are not inherently adversarial can be usefully modeled in this way, since the assumptions are sufficiently weak that effective prediction strategies for adversarial settings are very widely applicable.


  1. 1.
    Abernethy, J., Agarwal, A., Bartlett, P.L., Rakhlin, A.: A stochastic view of optimal regret through minimax duality. arXiv:0903.5328v1 [cs.LG] (2009)Google Scholar
  2. 2.
    Abernethy, J., Bartlett, P.L., Rakhlin, A., Tewari, A.: Optimal strategies and minimax lower bounds for online convex games. UC Berkeley EECS Technical Report EECS-2008-19 (2008)Google Scholar
  3. 3.
    Bartlett, P.L., Hazan, E., Rakhlin, A.: Adaptive online gradient descent. UC Berkeley EECS Technical Report EECS-2007-82 (2007)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Peter L. Bartlett
    • 1
  1. 1.Computer Science Division and Department of StatisticsUniversity of California at BerkeleyBerkeleyUSA

Personalised recommendations