Rianne van den Berg

Rianne van den Berg is a machine learning researcher who applies deep learning to computational chemistry at Microsoft Research AI for Science. She earned a PhD in theoretical physics at the University of Amsterdam in 2016, joined the Amsterdam Machine Learning Lab as a postdoctoral researcher, and won the Faculty of Science Lecturer of the Year award there in 2019. Her papers include Relational Graph Convolutional Networks in 2017 and discrete denoising diffusion models (D3PMs) in 2021, and she worked as a Research Scientist at Google Brain before joining Microsoft. In 2025, she coauthored Skala, a deep-learning exchange-correlation functional for density functional theory. She is a Senior Principal Research Manager who co-leads a large team of deep learning researchers, engineers and quantum chemists.

Now at: Microsoft
Born: 1988
Works: 39
Citations: 8,148
h-index: 18
Badges: Rising
Bridge
Links: Google Scholar, X and LinkedIn

Years link to their sources. Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

2016

A physics PhD in Amsterdam

She receives a PhD in theoretical physics at the University of Amsterdam in 2016. She then joins the university's Amsterdam Machine Learning Lab (AMLAB) as a postdoctoral researcher.
2017

Relational graph convolutional networks

In 2017, she coauthors Modeling Relational Data with Graph Convolutional Networks with Michael Schlichtkrull, Thomas Kipf, Peter Bloem, Ivan Titov and Max Welling. The paper introduces R-GCNs for knowledge base completion, and adding an R-GCN encoder to the DistMult model improves link prediction on FB15k-237 by 29.8% over a decoder-only baseline.
2019

Lecturer of the Year in Amsterdam

In 2019, she wins the Faculty of Science Lecturer of the Year award at the University of Amsterdam for teaching the core machine learning course in the master of AI.
2021

Google Brain and discrete diffusion

She works as a Research Scientist at Google Brain before joining Microsoft. In 2021, she coauthors Structured Denoising Diffusion Models in Discrete State-Spaces with Jacob Austin, Daniel D. Johnson, Jonathan Ho and Daniel Tarlow, which introduces D3PMs, diffusion-like generative models for discrete data.
2022

AI4Science at Microsoft Research Amsterdam

Microsoft announces a global AI for science team in July 2022, spanning the UK, China and the Netherlands. In December 2022, she is a Principal Researcher at Microsoft Research and presents the group's research areas in a seminar at EPFL, while Amsterdam Science Park reports that Max Welling started the Amsterdam lab together with her.
March 2024

Co-leading a density functional project

At the Microsoft Research Forum panel on transforming the natural sciences with AI in March 2024, she is a principal research manager in AI4Science at Microsoft Research in Amsterdam. She co-leads a project on density functional theory.
June 2025

Skala, a learned density functional

In June 2025, she coauthors the Microsoft Research blog post and the preprint that introduce Skala, a deep-learning exchange-correlation functional for density functional theory. Skala reaches chemical accuracy for atomization energies of small molecules at the computational cost typical of semi-local DFT.
As of October 2026

Senior Principal Research Manager

She is a Senior Principal Research Manager at Microsoft Research AI for Science, co-leading a large interdisciplinary team of deep learning researchers, engineers and quantum chemists. In August 2026, the team announces Skala 1.1, trained on 2.5 times more data than the first public version and available in CP2K.