MrPhilipp Wacker

Collaboration focus

“How can we reduce the number of physical experiments needed to answer this question?”

My research can help decide what experiment or measurement to run next, so they can learn faster with fewer trials. I use statistical modelling, Bayesian methods and optimisation to design experiments adaptively when experiments are expensive, slow or noisy.

Supervision

I'm looking for PhD, MMathSci, MSc, and Honours students in Maths&Stats wanting to do a project in 

• inference of continuous-time Markov chains
• optimal experimental design using copulas
• warding off under-/overfitting of mathematical models via hierarchical chain regularisation
• optimal sequential experimental design using Markov decision processes, with applications to epidemic management
• modelling mast seeding in trees from timeseries data

I currently have no funding opportunities available, but chat to me if you're interested! 

About me

I am a mathematician working at the interface of applied mathematics, statistics, and computational science. My research develops mathematical and statistical methods to extract reliable information from complex data and dynamical systems.

A central theme of my work is Bayesian inverse problems: how to infer unknown parameters, risks, or system states from indirect and noisy measurements. I design and analyse computational methods for uncertainty quantification, optimisation, and data assimilation, with particular expertise in particle-based algorithms and stochastic modelling.

I am especially interested in:

• Optimal experimental design: Developing principled methods to determine what data should be collected to maximise information gain while minimising cost.
• Digital twins: Building mathematically rigorous, data-driven virtual replicas of physical systems that integrate mechanistic models with real-time data for prediction, monitoring, and decision support.
• Optimisation and control of dynamical systems
• Statistical inference for complex and high-dimensional models

My work is motivated by applications in engineering, environmental systems, and biological processes. 

I welcome collaborations with partners interested in:

• Turning data into actionable insight
• Designing efficient experiments and monitoring strategies
• Developing predictive digital twins
• Quantifying and reducing uncertainty in complex systems