How Edge of Stability Hinders SCAFFOLD in Federated Optimization
In federated learning, it is well known that heterogeneous data can (in theory) slow down optimization, and much effort has been directed at designing optimization algorithms that are unaffected by data heterogeneity, such as the SCAFFOLD algorithm. Yet, despite strong theoretical guarantees, SCAFFOLD does not usually outperform the much simpler FedAvg in practice. In this work, we propose that this gap is due to the presence of Edge of Stability (EoS) and progressive sharpening in federated optimization, supported by extensive empirical probing. First, we find that EoS-like dynamics occur wit
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 51%Optimized IoT clustering and assignment in semi-synchronous federated learning - Nature →
- PossiblePossibly related (embedding) · 50%Long-horizon local optimization and regularized knowledge improve personalized federated recommendation - Bioengineer.org →
- FuzzyOverlapping authors or contributors · 62%modular/modular →
“Shared author/contributor keys: liu”
- LinkedLinked via arxiv author · 85%Anant Khandelwal →
“How Edge of Stability Hinders SCAFFOLD in Federated Optimization”
- LinkedLinked via arxiv author · 85%Michael Crawshaw →
“How Edge of Stability Hinders SCAFFOLD in Federated Optimization”
- LinkedLinked via arxiv author · 85%Mingrui Liu →
“How Edge of Stability Hinders SCAFFOLD in Federated Optimization”