Hamish Flynn

Hamish Flynn is a Researcher in the Department of Engineering Artificial Intelligence and Machine Learning. His work focuses on developing advanced algorithms for sequential decision-making systems, with a strong emphasis on theoretical guarantees and practical implementations.

His research interests span several core areas of machine learning:

• Bandit algorithms (contextual, linear, and multi-armed variants)
• PAC-Bayesian theory and applications
• Reinforcement learning systems
• Online optimization under uncertainty
• Statistical learning theory

Flynn's recent publications (2022-2025) demonstrate consistent focus on improving sequential learning methods. Key research themes include: non-iid noise modeling in bandits, confidence bound optimization for sequential regression, sparse nonparametric methods, and hardware-efficient machine learning implementations. His work frequently combines theoretical frameworks (PAC-Bayes, martingale theory) with practical applications in reinforcement learning and adaptive systems.

No scientific awards, students advised, grant activities, or lab affiliations are mentioned in the available information.