Spectral Sentinel: Scalable Byzantine-Robust Decentralized Federated Learning via Sketched Random Matrix Theory on Blockchain

TL;DR

Spectral Sentinel introduces a scalable, blockchain-integrated framework for Byzantine-robust decentralized federated learning, leveraging random matrix theory to detect malicious updates even in large models with heterogeneous data.

Contribution

It proposes a novel spectral detection method combining sketching and MP law tracking, enabling robust Byzantine defense on models up to 1.5 billion parameters with provable guarantees.

Findings

Achieves 78.4% accuracy against Byzantine attacks

Scales to models with 1.5 billion parameters

Outperforms baseline defenses in diverse attack scenarios

Abstract

Decentralized federated learning (DFL) enables collaborative model training without centralized trust, but it remains vulnerable to Byzantine clients that poison gradients under heterogeneous (Non-IID) data. Existing defenses face a scalability trilemma: distance-based filtering (e.g., Krum) can reject legitimate Non-IID updates, geometric-median methods incur prohibitive $O(n^2 d)$ cost, and many certified defenses are evaluated only on models below 100M parameters. We propose Spectral Sentinel, a Byzantine detection and aggregation framework that leverages a random-matrix-theoretic signature: honest Non-IID gradients produce covariance eigenspectra whose bulk follows the Marchenko-Pastur law, while Byzantine perturbations induce detectable tail anomalies. Our algorithm combines Frequent Directions sketching with data-dependent MP tracking, enabling detection on models up to 1.5B parameters using $O(k^2)$ memory with $k \ll d$. Under a $(\sigma,f)$ threat model with coordinate-wise honest variance bounded by $\sigma^2$ and $f < 1/2$ adversaries, we prove $(\epsilon,\delta)$-Byzantine resilience with convergence rate $O(\sigma f / \sqrt{T} + f^2 / T)$, and we provide a matching information-theoretic lower bound $\Omega(\sigma f / \sqrt{T})$, establishing minimax optimality. We implement the full system with blockchain integration on Polygon networks and validate it across 144 attack-aggregator configurations, achieving 78.4 percent average accuracy versus 48-63 percent for baseline methods.