CVSL Logo
FrancaisHome
AboutPeopleResearchPublicationsEventsProfile
About
Publications

 

 

 

REPARTI

MIVIM

Thompson Sampling for Combinatorial Bandits and its Application to Online Feature Selection


Audrey Durand and Christian Gagné


Abstract - In this work, we address the combinatorial optimization problem in the stochastic bandit setting with bandit feedback. We propose to use the seminal Thompson Sampling algorithm under an assumption on rewards expectations. More specifically, we tackle the online feature selection problem where results show that Thompson Sampling performs well. Additionnally, we discuss the challenges associated with online feature selection and highlight relevant future work directions.

download document

Bibtex:

@inproceedings{Durand1066,
    author    = { Audrey Durand and Christian Gagné },
    title     = { Thompson Sampling for Combinatorial Bandits and its Application to Online Feature Selection },
    booktitle = { Proceedings of the 28th AAAI Conference Workshop on Sequential Decision-Making with Big Data },
    pages     = { 6-9 },
    year      = { 2014 },
    month     = { July },
    location  = { Quebec City (QC), Canada }
}

Last modification: Jul 21 2014 11:10AM by audur2

     
   
   

©2002-. Computer Vision and Systems Laboratory. All rights reserved