mlsanityData to 7 Oct 2026

Feature Space Hijacking Attacks against Differentially Private Split Learning

arXiv 2201.04018 ↗First posted 11 Jan 2022cs.CR cs.AIcited by 14

Not on the map. Why?

Abstract (arXiv)

Split learning and differential privacy are technologies with growing potential to help with privacy-compliant advanced analytics on distributed datasets. Attacks against split learning are an important evaluation tool and have been receiving increased research attention recently. This work's contribution is applying a recent feature space hijacking attack (FSHA) to the learning process of a split neural network enhanced with differential privacy (DP), using a client-side off-the-shelf DP optimizer. The FSHA attack obtains client's private data reconstruction with low error rates at arbitrarily set DP epsilon levels. We also experiment with dimensionality reduction as a potential attack risk mitigation and show that it might help to some extent. We discuss the reasons why differential privacy is not an effective protection in this setting and mention potential other risk mitigation methods.

Key facts. Cited by 14 papers in the map (data through 7 Oct 2026). Builds on 11 of its 13 references that are in the map. First posted on arXiv 11 Jan 2022.

Builds on

11 of 13 references.

References · 11

  1. 2016Deep Learning with Differential Privacycited by 2,682
  2. 2017Differentially Private Federated Learning: A Client Level Perspectivecited by 451
  3. 2019On Evaluating Adversarial Robustnesscited by 442
  4. 2018Split learning for health: Distributed deep learning without sharing raw patient datacited by 391
  5. 2017Deep Models Under the GAN: Information Leakage from Collaborative Deep Learningcited by 326
  6. 2018Distributed learning of deep neural network over multiple agentscited by 277
  7. 2020Unleashing the Tiger: Inference Attacks on Split Learningcited by 104
  8. 2018A General Approach to Adding Differential Privacy to Iterative Training Procedurescited by 84
  9. 2021UnSplit: Data-Oblivious Model Inversion, Model Stealing, and Label Inference Attacks Against Split Learningcited by 52
  10. 2018No Peek: A Survey of private distributed deep learningcited by 30
  11. 2021SplitGuard: Detecting and Mitigating Training-Hijacking Attacks in Split Learningcited by 20

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 2; 2024: 3; 2025: 5; 2026: 1.

14 papers.

Sort:

  1. 2022Combined Federated and Split Learning in Edge Computing for Ubiquitous Intelligence in Internet of Things: State of the Art and Future Directionscited by 14
  2. 2022Differentially Private CutMix for Split Learning with Vision Transformercited by 13
  3. 2022Multi-Task Distributed Learning using Vision Transformer with Random Patch Permutationcited by 10
  4. 2023Secure Split Learning against Property Inference, Data Reconstruction, and Feature Space Hijacking Attackscited by 8
  5. 2024A Survey of Privacy Threats and Defense in Vertical Federated Learning: From Model Life Cycle Perspectivecited by 7
  6. 2024A Hybrid Quantum Neural Network for Split Learningcited by 5
  7. 2023SplitOut: Out-of-the-Box Training-Hijacking Detection in Split Learning via Outlier Detectioncited by 4
  8. 2025Adversarial Robustness of Partitioned Quantum Classifierscited by 2
  9. 2024Privacy-Preserving Collaborative Split Learning Framework for Smart Grid Load Forecastingcited by 1
  10. 2025A Taxonomy of Attacks and Defenses in Split Learningcited by 1
  11. 2025A Distributed Generative AI Approach for Heterogeneous Multi-Domain Environments under Data Sharing constraintscited by 1
  12. 2025Oops!... They Stole it Again: Attacks on Split Learningcited by 1
2 more
  1. 2025Server-Enforced Watermarking in U-Shaped Split Federated Learningcited by 0
  2. 2026LightSplit: Practical Privacy-Preserving Split Learning via Orthogonal Projectionscited by 0