mlsanityData to 7 Oct 2026

Semi-overlapping Multi-bandit Best Arm Identification for Sequential Support Network Learning

arXiv 2512.24959 ↗First posted 31 Dec 2025cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Many modern AI and ML problems require evaluating partners' contributions through shared yet asymmetric, computationally intensive processes and the simultaneous selection of the most beneficial candidates. Sequential approaches to these problems can be unified under a new framework, Sequential Support Network Learning (SSNL), in which the goal is to select the most beneficial candidate set of partners for all participants using trials; that is, to learn a directed graph that represents the highest-performing contributions. We demonstrate that a new pure-exploration model, the semi-overlapping multi-(multi-armed) bandit (SOMMAB), in which a single evaluation provides distinct feedback to multiple bandits due to structural overlap among their arms, can be used to learn a support network from sparse candidate lists efficiently. We develop a generalized GapE algorithm for SOMMABs and derive new exponential error bounds that improve the best known constant in the exponent for multi-bandit best-arm identification. The bounds scale linearly with the degree of overlap, revealing significant sample-complexity gains arising from shared evaluations. From an application point of view, this work provides a theoretical foundation and improved performance guarantees for sequential learning tools for identifying support networks from sparse candidates in multiple learning problems, such as in multi-task learning (MTL), auxiliary task learning (ATL), federated learning (FL), and in multi-agent systems (MAS).

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 35 of its 76 references that are in the map. First posted on arXiv 31 Dec 2025.

Builds on

35 of 76 references.

References · 35

  1. 2016Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimizationcited by 688
  2. 2019Which Tasks Should Be Learned Together in Multi-task Learning?cited by 367
  3. 2023Data Selection for Language Models via Importance Resamplingcited by 322
  4. 2021Efficiently Identifying Task Groupings for Multi-Task Learningcited by 218
  5. 2019Client-Edge-Cloud Hierarchical Federated Learningcited by 202
  6. 2021Monte Carlo Tree Search: A Review of Recent Modifications and Applicationscited by 158
  7. 2020A Survey on Negative Transfercited by 120
  8. 2022FedNest: Federated Bilevel, Minimax, and Compositional Optimizationcited by 60
  9. 2022Do Current Multi-Task Optimization Methods in Deep Learning Even Help?cited by 60
  10. 2022Towards a Standardised Performance Evaluation Protocol for Cooperative MARLcited by 48
  11. 2021Optimality and Stability in Federated Learning: A Game-theoretic Approachcited by 40
  12. 2021Game of Gradients: Mitigating Irrelevant Clients in Federated Learningcited by 35
23 more
  1. 2023BenchMARL: Benchmarking Multi-Agent Reinforcement Learningcited by 31
  2. 2019Top-k Combinatorial Bandits with Full-Bandit Feedbackcited by 28
  3. 2023ForkMerge: Mitigating Negative Transfer in Auxiliary-Task Learningcited by 28
  4. 2020Auxiliary Task Reweighting for Minimum-data Learningcited by 23
  5. 2013Learning Bayesian Network Structure from Massive Datasets: The "Sparse Candidate" Algorithmcited by 22
  6. 2019AutoSeM: Automatic Task Selection and Mixing in Multi-Task Learningcited by 22
  7. 2022Fairness in Federated Learning via Core-Stabilitycited by 22
  8. 2021Context-Aware Online Client Selection for Hierarchical Federated Learningcited by 20
  9. 2016Multi-Task Learning with Labeled and Unlabeled Taskscited by 17
  10. 2022PGFed: Personalize Each Client's Global Objective for Federated Learningcited by 13
  11. 2021Compositional federated learning: Applications in distributionally robust averaging and meta learningcited by 12
  12. 2022AANG: Automating Auxiliary Learningcited by 12
  13. 2024Free-Rider and Conflict Aware Collaboration Formation for Cross-Silo Federated Learningcited by 9
  14. 2023Client Selection for Generalization in Accelerated Federated Learning: A Multi-Armed Bandit Approachcited by 8
  15. 2023Federated Compositional Deep AUC Maximizationcited by 8
  16. 2023Multi-task Representation Learning for Pure Exploration in Linear Banditscited by 6
  17. 2023Multi-task Representation Learning for Pure Exploration in Bilinear Banditscited by 5
  18. 2023Multitask Learning with No Regret: from Improved Confidence Bounds to Active Learningcited by 4
  19. 2023Learning to Group Auxiliary Datasets for Moleculecited by 3
  20. 2024From Barlow Twins to Triplet Training: Differentiating Dementia with Limited Datacited by 3
  21. 2025Multi-objective methods in Federated Learning: A survey and taxonomycited by 3
  22. 2022A Coalition Formation Game Approach for Personalized Federated Learningcited by 2
  23. 2025AssayMatch: Learning to Select Data for Molecular Activity Modelscited by 1

Built on by

Not cited yet.