Source Anonymity for Private Random Walk Decentralized Learning

2025-05-11

AI Summary

In random-walk-based decentralized learning, model updates may expose the sender’s identity to receivers, posing a critical source-identifiability privacy risk. Method: We propose a privacy-preserving mechanism integrating public-key encryption with path anonymization. We formally define “source anonymity” and design a network-adaptive target-selection distribution that achieves provable anonymity on random regular graphs. Contribution/Results: Our theoretical analysis establishes that, under this mechanism, no receiver can distinguish any user as the sender; all users exhibit Ω(1/n)-level indistinguishability in terms of source likelihood. Compared to existing approaches, our method significantly enhances sender untraceability while maintaining communication efficiency and practical deployability. The scheme ensures strong anonymity guarantees without requiring trusted third parties or global synchronization, and is compatible with standard decentralized learning protocols.

Technology Category

• Machine Learning: Privacy
• Search and Optimization: Learning to Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Security and Privacy: Privacy-enhancing technologies
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

This paper considers random walk-based decentralized learning, where at each iteration of the learning process, one user updates the model and sends it to a randomly chosen neighbor until a convergence criterion is met. Preserving data privacy is a central concern and open problem in decentralized learning. We propose a privacy-preserving algorithm based on public-key cryptography and anonymization. In this algorithm, the user updates the model and encrypts the result using a distant user's public key. The encrypted result is then transmitted through the network with the goal of reaching that specific user. The key idea is to hide the source's identity so that, when the destination user decrypts the result, it does not know who the source was. The challenge is to design a network-dependent probability distribution (at the source) over the potential destinations such that, from the receiver's perspective, all users have a similar likelihood of being the source. We introduce the problem and construct a scheme that provides anonymity with theoretical guarantees. We focus on random regular graphs to establish rigorous guarantees.

Problem

• Ensuring source anonymity in decentralized random walk learning
• Designing privacy-preserving encryption for model updates
• Balancing source-destination probabilities for uniform anonymity

Innovation

• Public-key cryptography secures model updates
• Anonymization hides source identity effectively
• Network-dependent probability ensures uniform source likelihood