mlsanityData to 14 Oct 2026

Gaming and Cooperation in Federated Learning: What Can Happen and How to Monitor It

arXiv 2509.02391 ↗First posted 2 Sep 2025cs.LG cs.GT stat.MLcited by 0

Not on the map. Why?

Abstract (arXiv)

The success of federated learning (FL) ultimately depends on how strategic participants behave under partial observability, yet most formulations still treat FL as a static optimization problem. We instead view FL deployments as governed strategic systems and develop an analytical framework that separates welfare-improving behavior from metric gaming. Within this framework, we introduce indices that quantify manipulability, the price of gaming, and the price of cooperation, and we use them to study how rules, information disclosure, evaluation metrics, and aggregator-switching policies reshape incentives and cooperation patterns. We derive threshold conditions for deterring harmful gaming while preserving benign cooperation, and for triggering auto-switch rules when early-warning indicators become critical. Building on these results, we construct a design toolkit including a governance checklist and a simple audit-budget allocation algorithm with a provable performance guarantee. Simulations across diverse stylized environments and a federated learning case study consistently match the qualitative and quantitative patterns predicted by our framework. Taken together, our results provide design principles and operational guidelines for reducing metric gaming while sustaining stable, high-welfare cooperation in FL platforms.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 74 of its 103 references that are in the map. First posted on arXiv 2 Sep 2025.

Builds on

74 of 103 references.

References · 74

  1. 2016Deep Learning with Differential Privacycited by 2,728
  2. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,281
  3. 2020Extracting Training Data from Large Language Modelscited by 2,189
  4. 2016Equality of Opportunity in Supervised Learningcited by 2,007
  5. 2017Understanding Black-box Predictions via Influence Functionscited by 1,741
  6. 2016Concrete Problems in AI Safetycited by 1,642
  7. 2018Datasheets for Datasetscited by 1,386
  8. 2020WILDS: A Benchmark of in-the-Wild Distribution Shiftscited by 1,375
  9. 2018Model Cards for Model Reportingcited by 1,080
  10. 2020Beyond Accuracy: Behavioral Testing of NLP models with CheckListcited by 1,006
  11. 2018The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networkscited by 864
  12. 2017Counterfactual Fairnesscited by 804
38 more
  1. 2018How To Backdoor Federated Learningcited by 765
  2. 2018Byzantine-Robust Distributed Learning: Towards Optimal Statistical Ratescited by 711
  3. 2020RobustBench: a standardized adversarial robustness benchmarkcited by 661
  4. 2019Data Shapley: Equitable Valuation of Data for Machine Learningcited by 629
  5. 2018A Reductions Approach to Fair Classificationcited by 587
  6. 2018Exploiting Unintended Feature Leakage in Collaborative Learningcited by 520
  7. 2019Local Model Poisoning Attacks to Byzantine-Robust Federated Learningcited by 493
  8. 2018Spectral Signatures in Backdoor Attackscited by 484
  9. 2017Differentially Private Federated Learning: A Client Level Perspectivecited by 459
  10. 2021Dynabench: Rethinking Benchmarking in NLPcited by 459
  11. 2019Agnostic Federated Learningcited by 456
  12. 2019Fair Resource Allocation in Federated Learningcited by 394
  13. 2017Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairnesscited by 384
  14. 2018Client Selection for Federated Learning with Heterogeneous Resources in Mobile Edgecited by 352
  15. 2016Semi-supervised Knowledge Transfer for Deep Learning from Private Training Datacited by 348
  16. 2020FLTrust: Byzantine-robust Federated Learning via Trust Bootstrappingcited by 346
  17. 2018The Hidden Vulnerability of Distributed Learning in Byzantiumcited by 329
  18. 2020Performative Predictioncited by 327
  19. 2019Robust Aggregation for Federated Learningcited by 317
  20. 2019Differential Privacy Has Disparate Impact on Model Accuracycited by 316
  21. 2020Attack of the Tails: Yes, You Really Can Backdoor Federated Learningcited by 309
  22. 2019A Little Is Enough: Circumventing Defenses For Distributed Learningcited by 288
  23. 2019Towards Efficient Data Valuation Based on the Shapley Valuecited by 281
  24. 2018Scalable Private Learning with PATEcited by 269
  25. 2015Strategic Classificationcited by 257
  26. 2018Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shiftcited by 243
  27. 2020Differentially Private Learning Needs Better Features (or Much More Data)cited by 235
  28. 2018Multiaccuracy: Black-Box Post-Processing for Fairness in Classificationcited by 222
  29. 2018Amplification by Shuffling: From Local to Central Differential Privacy via Anonymitycited by 164
  30. 2018Mitigating Sybils in Federated Learning Poisoningcited by 161
  31. 2018Categorizing Variants of Goodhart's Lawcited by 149
  32. 2019Data Valuation using Reinforcement Learningcited by 116
  33. 2019Reward Tampering Problems and Solutions in Reinforcement Learning: A Causal Influence Diagram Perspectivecited by 108
  34. 2014Preserving Statistical Validity in Adaptive Data Analysiscited by 103
  35. 2020Stochastic Optimization for Performative Predictioncited by 93
  36. 2019Strategic Classification is Causal Modeling in Disguisecited by 92
  37. 2015Algorithmic Stability for Adaptive Data Analysiscited by 88
  38. 2021Proof-of-Learning: Definitions and Practicecited by 84

50 most cited of 74 All 74 (JSON)

Built on by

Not cited yet.