Peter Bartlett's Talks

• Large scale model selection and computational oracle inequalities [Slides: pdf]
  Conference on Statistical Learning and Data Mining Rackham Graduate School, University of Michigan, Ann Arbor, MI, June 5 - 7, 2012.
• Computational Oracle Inequalities for Large Scale Model Selection Problems [Slides: pdf]
  High-Dimensional Problems in Statistics Forschungsinstitut fur Mathematik, ETH Zurich, September 19-23, 2011.
• Online Prediction [Slides: pdf]
  Learning Theory: State of the Art Institut Henri Poincare, Paris, May 9-11, 2011.
• Optimal online prediction in adversarial environments [Slides: pdf]
  The Second Asian Conference on Machine Learning Tokyo Institute of Technology, Tokyo, Japan, November 8-10, 2010.
• An online allocation problem: Dark pools [Slides: pdf]
  The Mathematics of Ranking American Institute of Mathematics, Palo Alto, California, August 16-20, 2010.
• l1-regularized linear regression: persistence and oracle inequalities [Slides: pdf]
  Probability and Statistics - an international conference in honor of P.L. Hsu's 100th birthday Peking University, Beijing, China. July 6, 2010.
  10th International Vilnius Conference on Probability Theory and Mathematical Statistics. June 30, 2010.
• Convex methods for classification [Slides: pdf]
  IMS Medallion Lecture. June 2008.
• Optimism in Sequential Decision Making [Slides: pdf]
  UC Berkeley Statistics. September 2007.
• Consistency of AdaBoost [Slides: pdf]
  Google. May 2007.
• AdaBoost and other Large Margin Classifiers: Convexity in Classification [Slides: pdf]
  Presented at DASP 2006. December 2006.
• Convex methods for classification [Slides: pdf]
• AdaBoost and other Large Margin Classifiers: Convexity in Classification [Slides: ps]
  Presented at the Institute of Statistical Science, Academia Sinica, Taipei, Taiwan., July 31, 2006.
• AdaBoost is Universally Consistent [Slides: ps, pdf]
  Presented at the 2006 Summer Institute held by the Institute of Information Science (IIS), Academia Sinica, Taipei, Taiwan., August 3, 2006.
• Regression Methods for Pattern Classification: Statistical Properties of Large Margin Classifiers [Slides: ps, pdf]
  Presented at Mathematisches Forschungsinstitut Oberwolfach, October 16-22, 2005.
• Empirical Minimization and Risk Bounds [Slides: ps]
• Statistical Properties of Large Margin Classifiers [Slides: ps, pdf]
• Large Margin Classifiers: Convexity and Classification [Slides: ps, pdf]
• Large Margin Methods for Structured Classification: Exponentiated Gradient Algorithms [Slides: ps, ps.gz]
• Local Rademacher Averages and Empirical Minimization [Slides: ps, pdf]
• The Role of Convexity in Prediction Problems. [Slides: ps, ps.gz; Handouts: ps, ps.gz]
  Presented at UC Berkeley EECS Joint Colloquium Distinguished Lecture Series, September 17, 2003.
• Prediction Algorithms: Complexity, Concentration, and Convexity. [Slides: ps, ps.gz; Handouts: ps, ps.gz]
  Presented at SYSID2003: 13th IFAC Symposium on System Identification, Rotterdam, The Netherlands, 27-29 August, 2003.
  See: Extended abstract.
• Convexity, Classification, and Risk Bounds. [Slides: ps, ps.gz; Handouts: ps, ps.gz]
  Presented at Workshop on Advances in Machine Learning, Montreal, Canada, June 8-11, 2003, and AMS/IMS/SIAM Joint Summer Research Conference on Machine Learning, Statistics, and Discovery, Snowbird, Utah, June 22-26, 2003.
  See: Convexity, classification, and risk bounds. Peter L. Bartlett, Michael I. Jordan and Jon D. McAuliffe. Technical Report 638, Department of Statistics, U.C. Berkeley, 2003.
• NIPS'98 Tutorial (an introduction to learning theory)