analiza aiqbench
This research introduces BettiSplit, a framework that uses topological complexity metrics to identify and mitigate privacy leakage risks in split learning architectures. By pinpointing specific layers where feature inversion becomes feasible without relying on explicit attack simulations, the proposed method significantly enhances security against gradient leakage while maintaining model utility. The findings underscore the critical importance of analyzing intermediate representation regimes beyond simple architectural depth for building robust collaborative AI systems.
BettiSplit: Topology-Guided Privacy-Aware Split Learning Against Feature Inversion and Gradient Leakage
arXiv:2607.24556v1 Announce Type: cross Abstract: Split learning enables collaborative model training by partitioning neural networks across clients and servers. However, improper split placement can lead to severe privacy leakage through intermediate representations. In this work, we propose a topology-guided framework for privacy-aware split learning based on the persistent Betti complexity of smashed activations. Through comprehensive layer-wise analysis, we show that privacy risk in split learning is highly non-uniform across layers and exhibits sharp transition regions that are not captured by architectural depth alone. In particular, feature inversion fidelity increases from negligible reconstruction to as high as 0.98 SSIM at deeper, privacy-critical split points. We further demonstrate that Betti complexity consistently identifies representation regimes associated with elevated feature-space privacy leakage across architectures and datasets. Leveraging this observation, we introduce BettiSafe, a topology-guided split selection strategy that identifies privacy-sensitive layers without requiring explicit attack execution. BettiSafe improves resistance to feature inversion by 2 to 5 times compared to depth-based heuristics while preserving classification accuracy. In addition, Betti-based regularisation increases inversion difficulty by nearly 5 x without degrading model utility, enabling a favourable privacy utility tradeoff. Overall, our results highlight topological complexity as a promising structural descriptor for secure, adaptive, and representation-aware split learning in real-world collaborative systems
Abstract:Split learning enables collaborative model training by partitioning neural networks across clients and servers. However, improper split placement can lead to severe privacy leakage through intermediate representations. In this work, we propose a topology-guided framework for privacy-aware split learning based on the persistent Betti complexity of smashed activations. Through comprehensive layer-wise analysis, we show that privacy risk in split learning is highly non-uniform across layers and exhibits sharp transition regions that are not captured by architectural depth alone. In particular, feature inversion fidelity increases from negligible reconstruction to as high as 0.98 SSIM at deeper, privacy-critical split points. We further demonstrate that Betti complexity consistently identifies representation regimes associated with elevated feature-space privacy leakage across architectures and datasets. Leveraging this observation, we introduce BettiSafe, a topology-guided split selection strategy that identifies privacy-sensitive layers without requiring explicit attack execution. BettiSafe improves resistance to feature inversion by 2 to 5 times compared to depth-based heuristics while preserving classification accuracy. In addition, Betti-based regularisation increases inversion difficulty by nearly 5 x without degrading model utility, enabling a favourable privacy utility tradeoff. Overall, our results highlight topological complexity as a promising structural descriptor for secure, adaptive, and representation-aware split learning in real-world collaborative systems
Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
arXiv:2607.24556 [cs.LG]
(or arXiv:2607.24556v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2607.24556 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 301 cuvinte
Subiecte conexe
[AINews] not much happened today
Latent Space · 10 sept. 2026
TASG-Explore: Traversability-Aware Sector-Guided Exploration for Ground Robot on Uneven Terrain
arXiv · 9 sept. 2026
Don't Send What You Don't Need: Question-Guided Token Pruning as a Privacy Defense for Vision-Language Models
arXiv · 16 sept. 2026
Federated Semi-Supervised Graph Neural Networks with Prototype-Guided Pseudo-Labeling for Privacy-Preserving Gestational Diabetes Mellitus Prediction
arXiv · 30 sept. 2026