ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2202.04593
18
23

Stochastic Contextual Dueling Bandits under Linear Stochastic Transitivity Models

9 February 2022
Viktor Bengs
Aadirupa Saha
Eyke Hüllermeier
ArXivPDFHTML
Abstract

We consider the regret minimization task in a dueling bandits problem with context information. In every round of the sequential decision problem, the learner makes a context-dependent selection of two choice alternatives (arms) to be compared with each other and receives feedback in the form of noisy preference information. We assume that the feedback process is determined by a linear stochastic transitivity model with contextualized utilities (CoLST), and the learner's task is to include the best arm (with highest latent context-dependent utility) in the duel. We propose a computationally efficient algorithm, CoLSTIM\texttt{CoLSTIM}CoLSTIM, which makes its choice based on imitating the feedback process using perturbed context-dependent utility estimates of the underlying CoLST model. If each arm is associated with a ddd-dimensional feature vector, we show that CoLSTIM\texttt{CoLSTIM}CoLSTIM achieves a regret of order O~(dT)\tilde O( \sqrt{dT})O~(dT​) after TTT learning rounds. Additionally, we also establish the optimality of CoLSTIM\texttt{CoLSTIM}CoLSTIM by showing a lower bound for the weak regret that refines the existing average regret analysis. Our experiments demonstrate its superiority over state-of-art algorithms for special cases of CoLST models.

View on arXiv
Comments on this paper