Sybil-Poisoning Resilience in Decentralized Federated Learning With Verifiable Delay Puzzles

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

Decentralized Federated learning (DFL) enables collaborative machine learning with enhanced privacy by allowing participants to train models locally and share updates for aggregation instead of sharing raw data. However, such systems are vulnerable to poisoning attacks that may compromise the learning process. This threat becomes even more severe when combined with Sybil attacks, where adversaries contribute numerous malicious updates with minimal effort, amplifying their impact. To overcome these challenges, particularly in the permissionless setup, we propose SyDeLP, a Blockchain-enabled protocol. SyDeLP integrates Byzantine Tolerant Aggregation for poisoning mitigation with a novel Verifiable Delay Puzzle (VDP) to counter Sybil attacks. Honest behavior is incentivized by dynamically reducing puzzles difficulties, decreasing the computational burden for honest nodes over time. Theoretical analysis demonstrates the system's resilience to poisoning attacks and the security of VDPs against Sybil attacks. Empirical evaluation on two benchmark datasets, covering three types of poisoning attacks and two collusion strategies, shows that SyDeLP maintains poisoning resilience comparable to state-of-the-art defenses, despite operating in a permissionless environment.