Jonas Latz

Jonas Latz is a Lecturer in Applied Mathematics at The University of Manchester, Department of Mathematics. His research interests include Bayesian methods, stochastic optimization, and partial differential equations in data science. He has contributed to various areas such as deep learning, inverse problems, and uncertainty quantification. His work has been published in several peer-reviewed journals, including Inverse Problems, European Journal of Applied Mathematics, and Journal of Machine Learning Research. Latz has received several awards, including the SIAM Activity Group Uncertainty Quantification Early Career Prize in 2024 and the SIAM Student Paper Prize in 2020.

Jonas Latz is a researcher at the University of Manchester with expertise in Bayesian Inference, Numerical Analysis, Data Science, Uncertainty Quantification, and Machine Learning. His work focuses on developing and analyzing methods for Bayesian inverse problems, including well-posedness, sampling techniques, and uncertainty quantification. He has contributed to the application of stochastic gradient descent, multilevel sequential Monte Carlo methods, and data-driven approximation techniques. His research also explores the integration of physics-informed neural networks and Gaussian process priors in inverse modeling and image reconstruction. Latz's publications highlight the intersection of probabilistic modeling, computational methods, and scientific applications.