Daniel Commey

I develop methods for trustworthy distributed learning, focusing on verifiable evaluation, post-quantum authentication, and resilience to malicious participants. My work spans federated learning, IoT, and edge systems.

Prospective students: CSULB undergraduate and MS students interested in this work may email me at daniel.commey@csulb.edu. Please include a brief introduction, relevant coursework or project experience, and a research topic you would like to explore.

Research

Verifiable distributed learning

Zero-knowledge proofs that verify whether a local evaluation metric meets a threshold without revealing the metric.

Post-quantum security

Cryptographic foundations that prepare federated, decentralized, and edge systems for quantum threats.

Adversarial resilience

Defenses and incentives for systems in which participants may poison, manipulate, or game the protocol.

Selected Publications

JISA'26

PUFZIN: Secure and Scalable Blockchain-IoT with PUFs and Zero-Knowledge Proofs

Daniel Commey, S. G. Hounsinou, G. V. Crosby

Journal of Information Security and Applications, vol. 100, article 104510, 2026

Combines physical-unclonable-function device fingerprints with zero-knowledge proofs for scalable, privacy-preserving authentication in blockchain-IoT.

ESWA'26

PQS-BFL: A Post-Quantum Secure Blockchain-based Federated Learning Framework

Daniel Commey, G. V. Crosby

Expert Systems with Applications, 2026

Uses post-quantum authentication and blockchain verification to protect model updates in federated learning.

arXiv Preprint

ZKP-FedEval: Verifiable and Privacy-Preserving Federated Evaluation using Zero-Knowledge Proofs

Daniel Commey, B. Appiah, G. S. Klogo, G. V. Crosby

arXiv:2507.11649 [cs.LG], 2025

Verifies that a client’s local loss is below a threshold without revealing the loss value, using zero-knowledge proofs in federated evaluation.

JISA'25

Blockchain-Enabled Dynamic Honeypot Conversion for Resource-Efficient IoT Security

Daniel Commey, M. Nkoom, S. G. Hounsinou, G. V. Crosby

Journal of Information Security and Applications, 2025

Combines ML threat scoring and game-theoretic incentives to turn idle IoT devices into honeypots as threats change.

News