Nearly Tight Bounds for Cross-Learning Contextual Bandits with Graphical Feedback

Ruiyuan Huang, Zengfeng Huang · 2025

Read Nearly Tight Bounds for Cross-Learning Contextual Bandits with Graphical Feedback by Ruiyuan Huang, Zengfeng Huang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-04678.txt.

Description

Ruiyuan Huang and Zengfeng Huang study choice problems where each pick also teaches about linked picks, in a 2025 paper. A case in point is a first price auction, where one bid can hint at results for like bids. Past work left open if a key rate could be kept as cases grew.

The pair gives a method that meets that rate, even when losses are set by a foe and some arms give no direct news. The trick is to set one hard arm aside in each round and fix its drift. Tests on made up data back the claim, and the proof is checked in full by a proof tool.

Catalog record

CREATORRuiyuan Huang, Zengfeng Huang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE74 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-04678

Browse the full library · About the Tower of Babel