Material Detail

General Oracle Inequalities for Gibbs Posterior with Application to Ranking

This video was recorded at 26th Annual Conference on Learning Theory (COLT), Princeton 2013. In this paper, we summarize some recent results in Li et al. (2012), which can be used to extend an important PAC-Bayesian approach, namely the Gibbs posterior, to study the nonadditive ranking risk. The methodology is based on assumption-free risk bounds and nonasymptotic oracle inequalities, which leads to nearly optimal convergence rates and optimal model selection to balance the approximation errors and the stochastic errors.

Keywords:: videolectures, ocwc, oec

Disciplines:

Science and Technology / Computer Science / Programming & Programming Languages

Go to Material

Bookmark / Add to Course ePortfolio

Create a Learning Exercise

Add Accessibility Information

Rate

Add a Comment

Quality

User Rating
Comments
Learning Exercises
Bookmark Collections
Course ePortfolios
Accessibility Info

Report Broken Link
Report as Inappropriate

More about this material

Material Type:: Presentation
Date Added to MERLOT:: February 8, 2015
Date Modified in MERLOT:: February 8, 2015
Author:: Cheng Li, Department of Statistics, Northwestern University
Submitter:: The Open Education Consortium
Primary Audience:: College General Ed, College Lower Division, College Upper Division
Technical Format:: Video

Mobile Compatibility:: Not specified at this time
Language:: English
Cost Involved:: No
Source Code Available:: No
Accessibility Information Available:: Unknown
Creative Commons:: This work is licensed under a Attribution-NonCommercial-NoDerivs 3.0 United States