Weighted Sequential Bayesian Inference for Non-Stationary Linear Contextual Bandits

Name: Weighted Sequential Bayesian Inference for Non-Stationary Linear Contextual Bandits
Uploaded: 2026-08-20T14:15:53.643711+00:00
Duration: 653 s
Description: The hosts discuss 'Weighted Sequential Bayesian Inference for Non-Stationary Linear Contextual Bandits,' a paper providing a principled Bayesian framework for algorithms that learn in changing environments. They detail how the approach handles uncertainty, matches state-of-the-art performance, and offers practical improvements over existing methods.

The hosts discuss 'Weighted Sequential Bayesian Inference for Non-Stationary Linear Contextual Bandits,' a paper providing a principled Bayesian framework for algorithms that learn in changing environments. They detail how the approach handles uncertainty, matches state-of-the-art performance, and offers practical improvements over existing methods.

Open the full episode