Patrimony
The Louis Bachelier Group's patrimony has been defined as all the publications produced by academic researchers thanks to Group funding (ILB, FdR, IEF, Labex) or via the use of EquipEx data (BEDOFIH, EUROFIDAI).
Inference and applications for topic models.
Allocation de Dirichlet latente, Apprentissage en ligne, Apprentissage non supervisé, Determinantal point processes, Latent Dirichlet allocation, Latent variable models, Modèles thèmatiques, Modèles à variables latentes, Online learning, Processus ponctuels determinantaux, Topic models, Unsupervised learning
An online EM algorithm in hidden (semi-)Markov models for audio segmentation and clustering.
Audio segmentation, EM algorithm, Hidden Markov models, Online learning
Online but Accurate Inference for Latent Variable Models with Local Gibbs Sampling.
Gibbs Sampling, Latent Dirichlet Allocation, Latent Variables Models, Online Learning, Topic Modelling
On the optimality of the Hedge algorithm in the stochastic regime.
Adaptive algorithms, Hedge, Online learning, Prediction with expert advice
Portfolio choice, portfolio liquidation, and portfolio transition under drift uncertainty.
Bayesian learning, Hamilton–Jacobi–Bellman equations, Online learning, Optimal execution, Optimal portfolio choice, Optimal portfolio liquidation, Optimal portfolio transition, Stochastic optimal control
Stochastic tracking algorithms and empirical concentration inequalities for statistical learning.
Algorithmes Stochastiques, Apprentissage en Ligne, Bornes d'Erreur en Généralisation, Classification Multi-Classes, Empirical Bernstein Inequalities, Feature Selection, Generalization Bounds, Inégalités de Bernstein Empiriques, Martingales, Matching Pursuit, Multiclass Classification, Online Learning, Ranking, Stochastic Algorithms, Sélection de Caractéristiques, U-Statistics, U-Statistiques
Efficient online learning with kernels for adversarial large scale problems.
Kernel methods, Large scale, Online learning