Alireza Daneshkhah

Dr Alireza Daneshkhah holds a PhD (Warwick), is a fellow of the Royal Statistical Society and a member of International Society of Bayesian Analysis. He is a Senior Lecturer in Statistics and Course Director of Data Science and Computational Intelligence (MSc) in the School of Engineering, Environment and Computing of Coventry University. Alireza is Bayesian statistician interested in applying Bayesian probabilistic methods in a wide range of applications. These include modelling interdependencies of large scale data and simulation of complex systems using Bayesian networks, Gaussian process emulators and artificial neural networks. In the recent years, his research has been involved in developing Proactive asset management using optimised time-based maintenance and online condition based maintenance for networked infrastructure using advanced dynamic graphical models in the presence of massive heterogeneous information, including on-line data (SCADA and sensor data). He is also interested in probabilistic risk analysis of climate change affecting water systems and reliability analysis of the assets under natural hazard threats. His other research interests are using expert elicitation techniques when the available data is limited, and modelling Big data using a wide range of Machine learning techniques. Ali’s most recent research interest is to develop Deep learning methods using Deep Gaussian process and neural networks with a wide range of applications including uncertainty quantification of highly complex PDE based models, autonomous vehicle validation and testing, processing medical images, remote sensing, etc.

I teach various undergraduate and postgraduate modules on statistics, Artificial intelligence and Machine Learning and supervise PhD students both within CEM and jointly with other centres. My main focus is on topics in  Deep laerning using Deep Gaussain process, Probabilstic Uncertainty quantification and Sensitivity analysis, eliciting probability distributions, and modelling Big Data using Graphical models (particularly, Bayesian networks) and multivariate copulas (known as pair-copula vine).