Education Research Current About VU Amsterdam NL
Login as
Prospective student Student Employee
Bachelor Master VU for Professionals
Exchange programme VU Amsterdam Summer School Honours programme VU-NT2 Semester in Amsterdam
PhD at VU Amsterdam Research highlights Prizes and distinctions
Research institutes Our scientists Research Impact Support Portal Creating impact
News Events calendar Woman at the top
Israël and Palestinian regions Culture on campus
Practical matters Mission and core values Entrepreneurship on VU Campus
Organisation Partnerships Alumni University Library Working at VU Amsterdam
Sorry! De informatie die je zoekt, is enkel beschikbaar in het Engels.
This programme is saved in My Study Choice.
Something went wrong with processing the request.
Something went wrong with processing the request.

PhD defence D. Nieman 3 June 2025 09:45 - 11:15

Share
Asymptotics for variational Gaussian processes

Faster and More Reliable Statistics: New Research Makes Complex Data Analysis More Accessible

Analyzing large amounts of data using advanced statistical models is becoming significantly faster and more reliable—this is the outcome of recent research by mathematician Dennis Nieman. He explored how smart mathematical techniques—known as variational approaches—can make complex statistical calculations more efficient without sacrificing accuracy.

What does this mean in practice?
Statistical analysis, especially in science and technology, often demands extensive computing power and time. Think of models predicting the spread of diseases, the evolution of climate scenarios, or decision-making in self-driving cars. Bayesian methods, which explicitly account for uncertainty in their calculations, are well-suited for these tasks—but they’re also computationally intensive.

Nieman demonstrates that variational approaches—a kind of smart shortcut in the computational process—can significantly speed up these analyses. What’s crucial is that the method adapts well to the situation and accurately estimates uncertainty. His research now provides mathematical guarantees for when and how these approaches can be trusted.

Key insights from the research:

  • The dimension of the model—that is, the number of features it incorporates—is a critical factor in the quality of the outcome.
  • If the dimension is too low, the results become unreliable due to oversimplification.
  • If the model is sufficiently complex (high-dimensional), reliability improves, although it requires more computing power.
  • Nieman calculated the optimal balance between speed and accuracy for various methods.

Another important result: although variational methods are sometimes criticized for underestimating uncertainty, the models in this study show that it doesn’t have to be that way. With the right approach, these methods can indeed provide trustworthy estimates of uncertainty—an essential feature of any sound statistical analysis.

Why does this matter?
Whether you're working with medical data, climate models, or machine learning systems, these insights help researchers and data analysts make better choices in their statistical toolkit. Ultimately, users and consumers also benefit, as technology and science become faster, more efficient, and more reliable.

More information on the thesis

Programme

PhD defence by door D. Nieman

PhD Faculty of Science

Supervisors:

  • prof.dr.ir. F.H. van der Meulen
  • J. H. van Zanten
  • dr. B.T. Szabo

The PhD defence can be followed online as well

About PhD defence D. Nieman

Starting date

  • 3 June 2025

Time

  • 09:45 - 11:15

Location

  • VU Main Building

Address

  • De Boelelaan 1105
  • 1081 HV Amsterdam

Follow the defence online

Go to livestream

Quick links

Homepage Culture on campus VU Sports Centre Dashboard

Study

Academic calendar Study guide Timetable Canvas

Featured

VUfonds VU Magazine Ad Valvas Digital accessibility

About VU

Contact us Working at VU Amsterdam Faculties Divisions
Privacy Disclaimer Veiligheid Webcolofon Cookies Webarchief

Copyright © 2025 - Vrije Universiteit Amsterdam