mlsanityData to 7 Oct 2026

FedHypeVAE: Federated Learning with Hypernetwork Generated Conditional VAEs for Differentially Private Embedding Sharing

arXiv 2601.00785 ↗First posted 2 Jan 2026cs.LG cs.AI cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Federated data sharing promises utility without centralizing raw data, yet existing embedding-level generators struggle under non-IID client heterogeneity and provide limited formal protection against gradient leakage. We propose FedHypeVAE, a differentially private, hypernetwork-driven framework for synthesizing embedding-level data across decentralized clients. Building on a conditional VAE backbone, we replace the single global decoder and fixed latent prior with client-aware decoders and class-conditional priors generated by a shared hypernetwork from private, trainable client codes. This bi-level design personalizes the generative layerrather than the downstream modelwhile decoupling local data from communicated parameters. The shared hypernetwork is optimized under differential privacy, ensuring that only noise-perturbed, clipped gradients are aggregated across clients. A local MMD alignment between real and synthetic embeddings and a Lipschitz regularizer on hypernetwork outputs further enhance stability and distributional coherence under non-IID conditions. After training, a neutral meta-code enables domain agnostic synthesis, while mixtures of meta-codes provide controllable multi-domain coverage. FedHypeVAE unifies personalization, privacy, and distribution alignment at the generator level, establishing a principled foundation for privacy-preserving data synthesis in federated settings. Code: github.com/sunnyinAI/FedHypeVAE

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 34 of its 52 references that are in the map. First posted on arXiv 2 Jan 2026.

Builds on

34 of 52 references.

References · 34

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  4. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  5. 2018Federated Optimization in Heterogeneous Networkscited by 2,971
  6. 2017A Survey on Deep Learning in Medical Image Analysiscited by 1,182
  7. 2019Deep Leakage from Gradientscited by 1,004
  8. 2016HyperNetworkscited by 1,003
  9. 2020Inverting Gradients -- How easy is it to break privacy in federated learning?cited by 707
  10. 2017Differentially Private Federated Learning: A Client Level Perspectivecited by 451
  11. 2021Opacus: User-Friendly Differential Privacy Library in PyTorchcited by 375
  12. 2021Personalized Federated Learning using Hypernetworkscited by 210
22 more
  1. 2021Vision Transformers are Robust Learnerscited by 208
  2. 2021Evaluating Gradient Inversion Attacks and Defenses in Federated Learningcited by 196
  3. 2021Adversary Instantiation: Lower Bounds for Differentially Private Machine Learningcited by 193
  4. 2020Salvaging Federated Learning by Local Adaptationcited by 142
  5. 2020Provable Defense against Privacy Leakage in Federated Learning from Representation Perspectivecited by 94
  6. 2023Generative models improve fairness of medical classifiers under distribution shiftscited by 90
  7. 2022Auditing Privacy Defenses in Federated Learning via Generative Gradient Leakagecited by 72
  8. 2020A Framework for Evaluating Gradient Leakage Attacks in Federated Learningcited by 63
  9. 2021Personalized Federated Learning through Local Memorizationcited by 57
  10. 2021Privacy-preserving Federated Learning based on Multi-key Homomorphic Encryptioncited by 46
  11. 2022FedTP: Federated Learning by Transformer Personalizationcited by 36
  12. 2021PRECODE - A Generic Model Extension to Prevent Deep Gradient Leakagecited by 28
  13. 2022Personalization Improves Privacy-Accuracy Tradeoffs in Federated Learningcited by 23
  14. 2022FedIIC: Towards Robust Federated Learning for Class-Imbalanced Medical Image Classificationcited by 18
  15. 2022DPD-fVAE: Synthetic Data Generation Using Federated Variational Autoencoders With Differentially-Private Decodercited by 14
  16. 2023FedSoup: Improving Generalization and Personalization in Federated Learning via Selective Model Interpolationcited by 14
  17. 2023Share Your Representation Only: Guaranteed Improvement of the Privacy-Utility Tradeoff in Federated Learningcited by 13
  18. 2022SemiPFL: Personalized Semi-Supervised Federated Learning Framework for Edge Intelligencecited by 7
  19. 2022Privacy in Practice: Private COVID-19 Detection in X-Ray Images (Extended Version)cited by 4
  20. 2023Gradient Leakage Defense with Key-Lock Module for Federated Learningcited by 4
  21. 2024Privacy-preserving datasets by capturing feature distributions with Conditional VAEscited by 2
  22. 2025Embedding-Based Federated Data Sharing via Differentially Private Conditional VAEscited by 2

Built on by

Not cited yet.