5 papers
Consistency of variational approximations under bounded Kullback--Leibler divergence
Hien Duy Nguyen, Jacob Westerhout, Thomas Guilmeau +1
Variational methods are widely used to approximate posterior distributions in Bayesian inference when exact computation is infeasible. We study when such approximations inherit pos…
Convergence of projected stochastic natural gradient variational inference for various step size and sample or batch size schedules
Thomas Guilmeau, Hadrien Hendrikx, Florence Forbes
Stochastic natural gradient variational inference (NGVI) is a popular and efficient algorithm for Bayesian inference. Despite empirical success, the convergence of this method is s…
A divergence-based condition to ensure quantile improvement in black-box global optimization
Thomas Guilmeau, Emilie Chouzenoux, Víctor Elvira
Black-box global optimization aims at minimizing an objective function whose analytical form is not known. To do so, many state-of-the-art methods rely on sampling-based strategies…
Adaptive importance sampling for heavy-tailed distributions via -divergence minimization
Thomas Guilmeau, Nicola Branchini, Emilie Chouzenoux +1
Adaptive importance sampling (AIS) algorithms are widely used to approximate expectations with respect to complicated target probability distributions. When the target has heavy ta…
On variational inference and maximum likelihood estimation with the λ-exponential family
Thomas Guilmeau, Emilie Chouzenoux, Víctor Elvira
The λ-exponential family has recently been proposed to generalize the exponential family. While the exponential family is well-understood and widely used, this it not the case of t…