🔍 The Workshop Scope
As Distributed Machine Learning (ML) becomes the backbone of everything from healthcare to autonomous vehicles, we’ve reached a tipping point. Efficiency is no longer the only goal—resilience and security are now the primary bottlenecks. HotDiSec (Workshop on Hot Topics at the Intersection of Distributed Machine Learning and Security) bridges the gap between Distributed Systems, AI, and Cybersecurity. While other venues treat these fields in isolation, we focus on the “systemic friction” where they collide.
HotDiSec Pillars:
- Securing Distributed ML: Protecting Federated and Split Learning from poisoning and privacy leaks.
- Distributed ML for Security: Using edge intelligence to hunt for malware and anomalies.
- AI for System Defense: Using ML to orchestrate adaptive security protocols.
Inspired by “hot topic” venues that prioritize bold ideas over polished proofs, HotDiSec serves as a platform for high-impact position papers, preliminary breakthroughs, and—crucially—forensic “lessons learned” from failed security architectures. We believe that understanding why a secure distributed model failed in deployment is often more valuable than a theoretical success. Whether you are challenging long-held assumptions in Federated Learning, presenting novel attack vectors in Edge AI, or critiquing the current state of adversarial robustness, HotDiSec invites researchers and practitioners to join a candid, high-energy dialogue on the future of resilient, decentralized intelligence.