Ingo Steinwart

Ingo Steinwart is a mathematician and machine-learning researcher whose work spans support vector machines, statistical learning theory and cluster analysis. He earned a mathematics diploma at Oldenburg in 1997 and a doctorate at Jena in 2000, then worked in Jena and visited Mainz before joining Los Alamos in 2003. His 2002 Journal of Complexity paper addressed universal consistency for support vector machines, and his 2008 book with Andreas Christmann covered SVM classification and Gaussian-kernel methods. He published adaptive density-based clustering work in 2015, coauthored the liquidSVM report in 2017 and derived improved Gaussian-kernel SVM rates with Thomas Hamm for low-dimensional data. As of October 2026, he is a Unit Director and Fellow of ELLIS Unit Stuttgart and a professor of mathematics at the University of Stuttgart's Institute for Stochastics and Applications.

Now at: University of Stuttgart, International Max Planck Research School for Intelligent Systems and European Laboratory for Learning and Intelligent Systems
Born: 1970
Works: 365
Citations: 5,084
h-index: 35
Badges: Founder
Wayfarer

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

Timeline

1997–2000

Mathematics in Oldenburg and Jena

He earns a mathematics diploma at Carl-von-Ossietzky University in Oldenburg in 1997. He completes a mathematics doctorate at Friedrich-Schiller-University in Jena in 2000.
2000–2003

Research in Jena and Mainz

From 2000 to 2003, he works as a scientific staff member at Friedrich-Schiller-University in Jena. He visits Johannes-Gutenberg University in Mainz as a visiting scientist from March to September 2002 and publishes “Support vector machines are universally consistent” in Journal of Complexity in 2002.
2003–2010

Los Alamos and an SVM book

From March 2003 to April 2010, he works as a technical staff member in the CCS-3 group at Los Alamos National Laboratory; in July 2008, he becomes Scientist Level 4. With Andreas Christmann, he publishes Support Vector Machines in 2008, a book with chapters on SVM classification and Gaussian-kernel methods.
Since 2010

Stuttgart professor and UCSC adjunct

In April 2010, he becomes full professor at the University of Stuttgart's Institute for Stochastics and Applications in the Department of Mathematics. He serves as associate adjunct professor in the Department of Computer Science at UC Santa Cruz's Jack Baskin School of Engineering from January 2010 to June 2011.
2015

Adaptive density clustering

In 2015, he publishes “Fully Adaptive Density-Based Clustering” in Annals of Statistics. The paper presents an algorithm that estimates the smallest density level at which a distribution separates into multiple connected components.
Since 2017

liquidSVM and an IMPRS role

In July 2017, he joins the faculty of the International Max Planck Research School for Intelligent Systems in Stuttgart and Tübingen. With Philipp Thomann, he coauthors the technical report “liquidSVM: A Fast and Versatile SVM package,” describing a C++ package for classification and regression.
2020–2021

Adaptive rates for Gaussian SVMs

With Thomas Hamm, he derives improved regression and classification rates for Gaussian-kernel SVMs on data with low intrinsic dimension. The preprint appears in March 2020 and is last revised in April 2021.
As of October 2026

ELLIS Unit Director and Fellow

In 2021, he and Andreas Bulling found ELLIS Unit Stuttgart. As of October 2026, he is a Unit Director and Fellow of ELLIS, and a professor of mathematics at the University of Stuttgart's Institute for Stochastics and Applications. His research covers statistical learning theory, kernel-based learning algorithms and cluster analysis.