Bayesian Coresets: Revisiting the Nonconvex Optimization Perspective

Apr 14, 2021



Bayesian coresets have emerged as a promisingapproach for implementing scalable Bayesianinference. The Bayesian coreset problem in-volves selecting a (weighted) subset of the datasamples, such that the posterior inference us-ing the selected subset closely approximatesthe posterior inference using the full dataset.This manuscript revisits Bayesian coresetsthrough the lens of sparsity constrained opti-mization. Leveraging recent advances in ac-celerated optimization methods, we proposeand analyze a novel algorithm for coreset se-lection. We provide explicit convergence rateguarantees and present an empirical evalua-tion on a variety of benchmark datasets tohighlight our proposed algorithm’s superiorperformance compared to state-of-the-art onspeed and accuracy.



Über AISTATS 2021

The 24th International Conference on Artificial Intelligence and Statistics was held virtually from Tuesday, 13 April 2021 to Thursday, 15 April 2021.

Präsentation speichern

Soll diese Präsentation für 1000 Jahre gespeichert werden?

Wie speichern wir Präsentationen?

Ewigspeicher-Fortschrittswert: 0 = 0.0%


Empfohlene Videos

Präsentationen, deren Thema, Kategorie oder Sprecher:in ähnlich sind

Interessiert an Vorträgen wie diesem? AISTATS 2021 folgen