I am a Ph.D. student fascinated by machine learning and computational statistics. My research at Lancaster addresses predictive analytics and forecasting.

In my work, I am particularly looking into deep learning, probabilistic programming and its application in interdisciplinary fields such as neuroscience or business related-topics such connected vehicles.

Moreover, I spend attention to computational architectures such as cloud computing which enables to embed computationally intensive methods for high dimensional data into the business environment.

My general research interests lie in the areas of machine learning and computational statistics, and its application in neuroscience and operations research related application fields such as connected vehicles or energy markets.

The increasing computational performance of modern computation architectures as well as the volume, variety, velocity, and veracity of nowadays available data, foster the exploration and development of ways to automatically learn and understand complex representations of high dimensional data, such as time-dependent data from vehicle sensors or human brain signals. Learning such representations enables methodological improvements in various scientific fields and could lead into a new era of economic changes. This potential motivates my research interests in deep learning and probabilistic programming. Particularly the fusion of probabilistic approaches and deep learning fascinates me.

Therefore, I addressed deep learning and probabilistic programming in my master thesis, in which I aimed to develop a novel approach for predictive maintenance of connected vehicles in a big data architecture. In this work, I sought to consider not only algorithms but also the integration of data science from a system-level perspective. Regarding the practical application of machine learning and computational statistics, the algorithms itself are only a small fraction of a vast and technically complex architecture. Thus, distributed computing and cloud computing are in my scope of my research.

Currently, I focus on neuroimaging of human-brain signals, where I combine methods of signal processing, such as wavelets, and deep learning approaches. On the one hand, the human brain inspired a variety of machine learning algorithms, such as the first neural networks. On the other hand, machine learning plays a vital role in the discovery of the brain. The mutual exchange between both neuroscience and machine learning continuously stimulates both disciplines. Hence, neuroscience is not only an exciting application for machine learning, it also allows me to get a better understanding of the algorithms itself.

Moreover, my work is related to market segmentation of the airline market and I contribute to a demand management project in E-Fulfillment by providing a customer choice model.