Self-Creating Random Walks for Decentralized Learning under Pac-Man Attacks

2026-01-12

AI Summary

This work addresses the vulnerability of decentralized learning to “Pac-Man” attacks—in which malicious nodes silently drop visiting random walks, causing learning to stall—by proposing CREATE-IF-LATE (CIL), a fully decentralized resilient algorithm. CIL introduces, for the first time, a self-creating random walk mechanism that employs a self-triggered regeneration strategy to prevent walk extinction and ensure continuous learning. Theoretical analysis establishes the non-extinction, boundedness, and biased convergence of the walk population under this framework. Empirical evaluations on both synthetic and real-world datasets demonstrate that CIL effectively mitigates such attacks with only a linear-time overhead, substantially enhancing system robustness while maintaining learning progress.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Search and Optimization: Learning to Search
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Random walk (RW)-based algorithms have long been popular in distributed systems due to low overheads and scalability, with recent growing applications in decentralized learning. However, their reliance on local interactions makes them inherently vulnerable to malicious behavior. In this work, we investigate an adversarial threat that we term the ``Pac-Man''attack, in which a malicious node probabilistically terminates any RW that visits it. This stealthy behavior gradually eliminates active RWs from the network, effectively halting the learning process without triggering failure alarms. To counter this threat, we propose the CREATE-IF-LATE (CIL) algorithm, which is a fully decentralized, resilient mechanism that enables self-creating RWs and prevents RW extinction in the presence of Pac-Man. Our theoretical analysis shows that the CIL algorithm guarantees several desirable properties, such as (i) non-extinction of the RW population, (ii) almost sure boundedness of the RW population, and (iii) convergence of RW-based stochastic gradient descent even in the presence of Pac-Man with a quantifiable deviation from the true optimum. Moreover, the learning process experiences at most a linear time delay due to Pac-Man interruptions and RW regeneration. Our extensive empirical results on both synthetic and public benchmark datasets validate our theoretical findings.

Problem

• decentralized learning
• random walks
• Pac-Man attack
• malicious nodes
• resilience

Innovation

• self-creating random walks
• Pac-Man attack
• decentralized learning
• resilient algorithm
• random walk extinction