Enhancing Federated Quadruplet Learning: Stochastic Client Selection and Embedding Stability Analysis
Not on the map. Why?
Abstract (arXiv)
Federated Learning (FL) enables decentralised model training across distributed clients without requiring data centralisation. However, the generalisation performance of the global model is usually degraded by data heterogeneity across clients, particularly under limited data availability and class imbalance. To address this challenge, we propose FedQuad, a novel method that explicitly enforces minimising intra-class representations while enabling inter-class splits across clients. By jointly minimising distances between positive pairs and maximising distances between negative pairs, the proposed approach mitigates representation misalignment introduced during model aggregation. We evaluate our method on CIFAR-10, CIFAR-100, and Tiny-ImageNet under diverse non-IID settings and varying numbers of clients, demonstrating consistent improvements over existing baselines. Additionally, we provide a comprehensive analysis of metric learning-based approaches in both centralised and federated environments, highlighting their effectiveness in alleviating representation collapse under heterogeneous data distributions.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 28 of its 36 references that are in the map. First posted on arXiv 8 May 2026.
Builds on
28 of 36 references.
References · 28
- 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
- 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,152
- 2020Supervised Contrastive Learningcited by 2,829
- 2021Model-Contrastive Federated Learningcited by 766
- 2014Deep metric learning using Triplet networkcited by 422
- 2021Federated Learning on Non-IID Data: A Surveycited by 346
- 2022FedDC: Federated Learning with Non-IID Data via Local Drift Decoupling and Correctioncited by 181
- 2019SCAFFOLD: Stochastic Controlled Averaging for Federated Learningcited by 175
- 2017Beyond triplet loss: a deep quadruplet network for person re-identificationcited by 172
- 2022Federated Learning from Pre-Trained Models: A Contrastive Learning Approachcited by 142
- 2021Local Learning Matters: Rethinking Data Heterogeneity in Federated Learningcited by 114
- 2020Hard negative examples are hard, but usefulcited by 82
16 more
- 2021FedProc: Prototypical Contrastive Federated Learning on Non-IID datacited by 81
- 2022Tackling Data Heterogeneity in Federated Learning with Class Prototypescited by 67
- 2022Orchestra: Unsupervised Federated Learning via Globally Consistent Clusteringcited by 45
- 2022Federated Minimax Optimization: Improved Convergence Analyses and Algorithmscited by 37
- 2021Federated Contrastive Learning for Decentralized Unlabeled Medical Imagescited by 34
- 2024Relaxed Contrastive Learning for Federated Learningcited by 32
- 2022Rethinking Data Heterogeneity in Federated Learning: Introducing a New Notion and Standard Benchmarkscited by 24
- 2022Vision Transformers in 2022: An Update on Tiny ImageNetcited by 13
- 2024Rethinking the Representation in Federated Unsupervised Learning with Non-IID Datacited by 13
- 2025FedRSClip: Federated Learning for Remote Sensing Scene Classification Using Vision-Language Modelscited by 11
- 2024A Mutual Information Perspective on Federated Contrastive Learningcited by 7
- 2023FedTrip: A Resource-Efficient Federated Learning Method with Triplet Regularizationcited by 6
- 2025Robust Asymmetric Heterogeneous Federated Learning with Corrupted Clientscited by 5
- 2024A New Theoretical Perspective on Data Heterogeneity in Federated Optimizationcited by 4
- 2025Tackling Data Heterogeneity in Federated Learning through Knowledge Distillation with Inequitable Aggregationcited by 1
- 2025FedQuad: Federated Stochastic Quadruplet Learning to Mitigate Data Heterogeneitycited by 1
Built on by
Not cited yet.