# Adaptive Federated Learning Defences via Trust-Aware Deep Q-Networks

arXiv: https://arxiv.org/abs/2510.01261 · HTML: https://mlsanity.com/paper/2510.01261/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Federated unlearning and poisoning, one of 12 directions in LLM safety and security.
- Builds on 10 of its 17 references that are in the map.
- First posted on arXiv 25 Sep 2025.

## Placed in

- LLM safety and security / [Federated unlearning and poisoning](https://mlsanity.com/safety/federated-unlearning-and-poisoning/)

## Builds on (most cited first, 10 of 10)

- [Communication-Efficient Learning of Deep Networks from Decentralized Data](https://mlsanity.com/paper/1602.05629/) (2016)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Local Model Poisoning Attacks to Byzantine-Robust Federated Learning](https://mlsanity.com/paper/1911.11815/) (2019)
- [Analyzing Federated Learning through an Adversarial Lens](https://mlsanity.com/paper/1811.12470/) (2018)
- [A Little Is Enough: Circumventing Defenses For Distributed Learning](https://mlsanity.com/paper/1902.06156/) (2019)
- [Blind Backdoors in Deep Learning Models](https://mlsanity.com/paper/2005.03823/) (2020)
- [A Reputation Mechanism Is All You Need: Collaborative Fairness and Adversarial Robustness in Federated Learning](https://mlsanity.com/paper/2011.10464/) (2020)
- [Model Poisoning Attacks to Federated Learning via Multi-Round Consistency](https://mlsanity.com/paper/2404.15611/) (2024)
- [Multi-Task Federated Reinforcement Learning with Adversaries](https://mlsanity.com/paper/2103.06473/) (2021)
- [Integrating Identity-Based Identification against Adaptive Adversaries in Federated Learning](https://mlsanity.com/paper/2504.03077/) (2025)

## Built on by (most cited first, 1 of 1)

- [Federated Client Selection under Partial Visibility: A POMDP Approach with Spatio-Temporal Attention](https://mlsanity.com/paper/2605.11752/) (2026)

## Abstract (arXiv)

Federated learning is vulnerable to poisoning and backdoor attacks under partial observability. We formulate defence as a partially observable sequential decision problem and introduce a trust-aware Deep Q-Network that integrates multi-signal evidence into client trust updates while optimizing a long-horizon robustness--accuracy objective. On CIFAR-10, we (i) establish a baseline showing steadily improving accuracy, (ii) show through a Dirichlet sweep that increased client overlap consistently improves accuracy and reduces ASR with stable detection, and (iii) demonstrate in a signal-budget study that accuracy remains steady while ASR increases and ROC-AUC declines as observability is reduced, which highlights that sequential belief updates mitigate weaker signals. Finally, a comparison with random, linear-Q, and policy gradient controllers confirms that DQN achieves the best robustness--accuracy trade-off.