SIERRA - 2019 - Annual activity report

SIERRA

SIERRA - 2019

Project-Team Sierra

Team, Visitors, External Collaborators

Overall Objectives

Statement

Research Program

Supervised Learning
Unsupervised Learning
Parsimony
Optimization

Application Domains

Applications for Machine Learning

Highlights of the Year

New Software and Platforms

ProxASAGA
object-states-action

New Results

Fast Gradient Methods for Symmetric Nonnegative Matrix Factorization.
Naive Feature Selection: Sparsity in Naive Bayes.
Regularity as Regularization: Smooth and Strongly Convex Brenier Potentials in Optimal Transport.
Ranking and synchronization from pairwise measurements via SVD.
Polyak Steps for Adaptive Fast Gradient Methods.
An Accelerated Decentralized Stochastic Proximal Algorithm for Finite Sums.
On Lazy Training in Differentiable Programming.
Implicit Regularization of Discrete Gradient Dynamics in Linear Neural Networks.
Efficient Primal-Dual Algorithms for Large-Scale Multiclass Classification.
Fast and Faster Convergence of SGD for Over-Parameterized Models (and an Accelerated Perceptron).
Globally Convergent Newton Methods for Ill-conditioned Generalized Self-concordant Losses.
Efficient online learning with kernels for adversarial large scale problems
Affine Invariant Covariance Estimation for Heavy-Tailed Distributions
Beyond Least-Squares: Fast Rates for Regularized Empirical Risk Minimization through Self-Concordance
Statistical Estimation of the Poincaré constant and Application to Sampling Multimodal Distributions
Stochastic first-order methods: non-asymptotic and computer-aided analyses via potential functions
Optimal Complexity and Certification of Bregman First-Order Methods
Efficient First-order Methods for Convex Minimization: a Constructive Approach

Bilateral Contracts and Grants with Industry

Bilateral Contracts with Industry
Bilateral Grants with Industry

Partnerships and Cooperations

National Initiatives
European Initiatives
International Research Visitors

Dissemination

Promoting Scientific Activities
Teaching - Supervision - Juries
Popularization

Bibliography

Publications of the year

Previous |

Home | Next next

Section: Research Program

Supervised Learning

This part of our research focuses on methods where, given a set of examples of input/output pairs, the goal is to predict the output for a new input, with research on kernel methods, calibration methods, and multi-task learning.

Previous |

Home | Next next