[PDF][PDF] HWA: Hyperparameters Weight Averaging in Bayesian Neural Networks

B Karimi, P Li - 3rd Symposium on Advances in Approximate Bayesian …, 2021 - hal.science
Bayesian neural networks attempt to combine the strong predictive performance of neural
networks with formal quantification of uncertainty of the predicted output in the Bayesian …

Non-Convex Optimization for Latent Data Models: Algorithms, Analysis and Applications

B Karimi - 2019 - theses.hal.science
Many problems in machine learning pertain to tackling the minimization of a possibly non-
convex and non-smooth function defined on a Many problems in machine learning pertain to …

[PDF][PDF] Nonconvex Optimization for Latent Data Models: An Incremental and An Online Point of View

B Karimi - 2019 - belhalk.github.io
• Models where the input-output relationship is not completely characterized by the observed
(x, y)∈ X× Y pairs in the training set• Dependence on a set of unobserved latent variables …