English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

An LP-Based Sampling Policy for Multi-Armed Bandits with Side-Observations and Stochastic Arm Availability (UCB-LP-A)

Forum topic · 小凯 · 2026-03-31

Summary

This paper (arXiv:2503.23700, by Ashutosh Soni, Peizhong Ju, and Atilla Eryilmaz, posted 2025-03-30) studies the stochastic multi-armed bandit (MAB) problem with side-observations over an underlying network structure. A bipartite graph connects actions to unknowns, so pulling an action reveals observations of all linked unknowns. Unlike prior work that assumes all arms are permanently accessible, the authors consider a more realistic setting where the set of playable arms (the activation set) changes randomly each round, modeling real systems such as social networks where peer side-information is available only intermittently. They propose UCB-LP-A, a policy that uses linear programming to compute the optimal sampling distribution over each achievable activation set, ensuring necessary observations are collected only via currently active arms while balancing exploration and exploitation. The paper derives theoretical regret upper bounds that characterize the influence of network structure and activation probabilities, and numerical simulations show UCB-LP-A significantly outperforms existing heuristics that ignore side-information or availability constraints.

Paper Overview

  • Field: Machine Learning
  • Authors: Ashutosh Soni, Peizhong Ju, Atilla Eryilmaz
  • Posted: 2025-03-30
  • arXiv: 2503.23700
  • Summary

    We study the stochastic multi-armed bandit (MAB) problem where an underlying network structure enables side-observations across related actions. A bipartite graph connects actions to a set of unknowns, so selecting an action reveals observations of all unknowns connected to it. While previous work assumes all actions are permanently accessible, we study the more practical stochastic availability setting, where the set of playable arms (the "activation set") changes dynamically each round. This framework models real systems with structural dependencies and volatility, such as social networks, where users provide side-information about peer preferences but are not always online and queryable.

    To address this challenge, we propose UCB-LP-A, a novel policy that uses a linear programming (LP) approach to optimize the exploration-exploitation trade-off under stochastic availability. Unlike standard network bandit algorithms that assume constant access, UCB-LP-A computes the optimal sampling distribution achievable on each activation set, ensuring the necessary observations are collected using only the currently active arms.

    Contributions

  • Models MAB with side-observations via a bipartite graph under stochastic arm availability (varying activation sets per round).
  • Proposes UCB-LP-A, an LP-based sampling policy for the exploration-exploitation trade-off under availability constraints.
  • Derives theoretical regret upper bounds characterizing the impact of network structure and activation probabilities.
  • Numerical simulations show UCB-LP-A significantly outperforms existing heuristics that ignore side-information or availability constraints.
--- *Auto-collected on 2026-03-31*

Tags

#multi-armed-bandits#reinforcement-learning#linear-programming#regret-analysis#side-observations#stochastic-availability#arxiv#paper

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177169451