معرفی
Omar Rivasplata is a Senior Lecturer in Machine Learning and Robotics, specializing in theoretical and applied machine learning. He is affiliated with the MCAIF: Centre for AI Fundamentals, focusing on advancing foundational aspects of artificial intelligence. His research spans reinforcement learning, Bayesian analysis, neural networks, and generalization bounds.
Key research interests include the theoretical underpinnings of deep learning architectures, optimization strategies for large-scale models, and probabilistic methods in machine learning. Notable contributions include work on gradient clipping for wide/deep networks and semi-pessimistic reinforcement learning frameworks.
- Leading the MCAIF project (2021–2026), a multidisciplinary initiative exploring core AI principles.
- Active collaborations with institutions worldwide, emphasizing open-access publications (e.g., Transactions on Machine Learning Research).
His articles explore topics like PAC-Bayesian theory, convergence of diffusion models, and meta-analyses of Bayesian methods. He advises a diverse cohort of postgraduate researchers in AI fundamentals.
