CLASSIC - 2012 - Annual activity report

CLASSIC

CLASSIC - 2012

Project-Team Classic

Members

Overall Objectives

Scientific Foundations

Regression models of supervised learning
On-line aggregation of predictors for the prediction of time series, with or without stationarity assumptions
Multi-armed bandit problems, prediction with limited feedback

Application Domains

Forecasting of the electricity consumption
Forecasting of the air quality
Forecasting of the production data of oil reservoirs
Data mining, massive data sets
Computational linguistics
Statistical inference on biological data

New Results

Contributions earlier to 2012 but only published in 2012
Extended journal versions written in 2012 of conference papers published in 2011
Bayesian methods
Sequential learning
Regression, classification, regression methods
Sparsity and $ℓ_{1}$ –regularization
Computational linguistics

Bilateral Contracts and Grants with Industry

Students Paid by Industrial Partners / Contracts with Industry

Partnerships and Cooperations

National Initiatives
European Initiatives
International Initiatives

Dissemination

Scientific Animation
Teaching - Supervision - Juries
Popularization

Bibliography

Major publications
Publications of the year

Previous |

Home | Next next

Section: New Results

Extended journal versions written in 2012 of conference papers published in 2011

Participants : Sébastien Gerchinovitz, Gilles Stoltz.

We wrote extended journal papers of some conference papers discussed in previous annual activity reports; they correspond to refences [32] , [19] , [20] .

Previous |

Home | Next next