University of Surrey 2022 Uncertainty Quantification for Robust AI through Optimal Transport

Machine learning has been making decisions that affect our lives. Yet, we often cannot even tell whether they are uncertain about their decisions. In this project, we will develop Bayesian techniques with tools from the optimal transport theory to better represent and quantify uncertainties in machine learning models. While theoretical results are promising, the deployment of the optimal transport theory in a wide range of machine learning applications is limited due to its heavy computational burden. We will derive algorithms for uncertainty propagation and quantification based on computationally efficient approximate optimal transport methods. The resulted toolkit will be validated on a real-world clinical application and is transferable across a wide range of safety-critical AI applications.   The successful applicant will be supervised by Dr Yunpeng Li and co-supervised by Professor Wenwu Wang. The PhD student will be based at the Nature Inspired Computing and Engineering (NICE) research group in the Department of Computer Science at the University of Surrey. The student will also benefit from resources from the Centre for Vision, Speech and Signal Processing in the Department of Electrical and Electronic Engineering at the University of Surrey. 

University of Surrey 2022 Uncertainty Quantification for Robust AI through Optimal Transport
Machine learning has been making decisions that affect our lives. Yet, we often cannot even tell whether they are uncertain about their decisions. In this project, we will develop Bayesian techniques with tools from the optimal transport theory to better represent and quantify uncertainties in machine learning models. While theoretical results are promising, the deployment of the optimal transport theory in a wide range of machine learning applications is limited due to its heavy computational burden. We will derive algorithms for uncertainty propagation and quantification based on computationally efficient approximate optimal transport methods. The resulted toolkit will be validated on a real-world clinical application and is transferable across a wide range of safety-critical AI applications.   The successful applicant will be supervised by Dr Yunpeng Li and co-supervised by Professor Wenwu Wang. The PhD student will be based at the Nature Inspired Computing and Engineering (NICE) research group in the Department of Computer Science at the University of Surrey. The student will also benefit from resources from the Centre for Vision, Speech and Signal Processing in the Department of Electrical and Electronic Engineering at the University of Surrey.