Rob Cornish

Dr. Rob Cornish is a Florence Nightingale Bicentenary Research Fellow at the University of Oxford's Department of Statistics. His research focuses on robustness in complex systems, emphasizing valid methodologies under minimal assumptions for real-world applications. He holds a DPhil from the AIMS CDT program under Arnaud Doucet and George Deligiannidis, and conducted postdoctoral work with Chris Holmes and Arnaud Doucet on causal inference and conformal prediction.

Key research areas include uncertainty quantification, causal inference, and machine learning. His work spans Bayesian computational methods, Monte Carlo techniques, and generative models. Recent publications highlight advancements in neural network symmetries, equivariant diffusions, and causal falsification of digital twins. He is affiliated with the Statistical Theory and Methodology and Computational Statistics and Machine Learning research groups.

Publications reflect a trend toward integrating theoretical rigor with applied challenges, such as scalable Bayesian inference and robust predictive models. His contributions bridge foundational statistical theory and modern machine learning applications, particularly in safety-critical system reliability.