Reduced kernel rules for classification

Add to your list(s) Download to your calendar using vCal

• Frederic Desobry, Sigproc. Lab. CUED
• Thursday 29 March 2007, 13:00-14:00
• LR5, Engineering, Department of.

If you have a question about this talk, please contact Taylan Cemgil.

A new methodology is proposed to discriminate between two probability measures known through a set of data distributed according to either of these two measures. A decision rule is built as the plug-in of a kernel rule, defined on a small subset of the learning set. This methodology allows for fast yet accurate estimates of the optimal classification rule. A statistical analysis yields consistency results, and rates of convergence for the probability of error. A dedicated model selection procedure is described, and experiments illustrate further the comparison to state-of-the art classifiers.

This talk is part of the Signal Processing and Communications Lab Seminars series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• Cambridge Centre for Data-Driven Discovery (C2D3)
• Cambridge University Engineering Department Talks
• Cambridge talks
• Centre for Smart Infrastructure & Construction
• Chris Davis' list
• Computational Continuum Mechanics Group Seminars
• Featured lists
• Information Engineering Division seminar list
• Interested Talks
• LR5, Engineering, Department of
• School of Technology
• Signal Processing and Communications Lab Seminars
• Trust & Technology Initiative - interesting events
• bld31
• ndk22's list
• ob366-ai4er
• rp587

Note that ex-directory lists are not shown.