Marc Deisenroth
Also known as Marc Peter Deisenroth
Marc Deisenroth is a machine learning researcher whose work spans data-efficient learning, probabilistic modeling, and autonomous decision-making. He earned a computer-science Diplom (MSc) in 2006 and a machine-learning PhD in 2009; in 2011, he co-authored PILCO, a data-efficient method for model-based policy search. At Imperial, he helped launch the Machine Learning Initiative in 2017, directed its Machine Learning Lab from 2018 to 2019, and received the President’s Award for Outstanding Early Career Researcher in 2018. He took up the Google DeepMind Chair of Machine Learning and Artificial Intelligence at UCL in 2019, co-authored Mathematics for Machine Learning in 2020, and shared an ICML 2020 Outstanding Paper honorable mention for Gaussian-process sampling work. From January 2024 to January 2025, he directed the Environment and Sustainability Grand Challenge at The Alan Turing Institute; since February 2025, he has been a Research Director at Google DeepMind and remains UCL’s Google DeepMind Chair.
- Now at
- University College London and Google DeepMind
- Works
- 219
- Citations
- 12,131
- h-index
- 41
Publication numbers from OpenAlex, as of 15 October 2026.
Timeline
-
2000–2009
Studying computer science and machine learning
From 2000 to 2006, he studies computer science at Universität Karlsruhe and earns a Diplom (MSc). He completes a machine-learning PhD at Karlsruhe Institute of Technology in 2009.
-
2010–2013
PILCO and early research posts
From February 2010 to December 2011, he works as a Research Associate at the University of Washington. In 2011, he and Carl Edward Rasmussen introduce PILCO, a model-based policy-search method that uses probabilistic dynamics to learn with little data. From December 2011 to August 2013, he leads Learning for Control at TU Darmstadt, then joins Imperial College London as a Research Fellow in September 2013.
-
2017
Imperial's Machine Learning Initiative
In 2017, he and colleagues launch Imperial's Machine Learning Initiative to bring machine learning researchers together for teaching and research.
-
2018
Director of Imperial's Machine Learning Lab
He becomes Director of Imperial's Machine Learning Lab in 2018, a role he holds through 2019. Imperial gives him the President's Award for Outstanding Early Career Researcher in 2018.
-
2019
Google DeepMind Chair at UCL
In October 2019, he takes up the Google DeepMind Chair of Machine Learning and Artificial Intelligence in UCL's Department of Computer Science.
-
2020
A textbook and ICML award
In 2020, he co-authors Mathematics for Machine Learning with A. Aldo Faisal and Cheng Soon Ong; Cambridge publishes it as a textbook on mathematical tools for machine learning. His five-author paper Efficiently sampling functions from Gaussian process posteriors receives an Outstanding Paper honorable mention at ICML 2020 and proposes scalable posterior sampling by separating the prior from the data.
-
Since 2025
Research Director at Google DeepMind
From January 2024 to January 2025, he directs the Environment and Sustainability Grand Challenge at The Alan Turing Institute. In February 2025, he joins Google DeepMind as a Research Director and continues as the Google DeepMind Chair of Machine Learning and Artificial Intelligence at UCL. His research spans data-efficient machine learning, probabilistic modeling and autonomous decision-making.
Sources and credits
Photo: We have no published photo of Marc Deisenroth yet. Send us a free-licensed photo.
- Curriculum Vitae | Marc Deisenroth
- Marc Deisenroth | About | Imperial College London
- PILCO: A Model-Based and Data-Efficient Approach to Policy Search
- Machine learning research: the driving force of Artificial Intelligence | Imperial News | Imperial College London
- Mathematics for Machine Learning | Cambridge Aspire website
- Efficiently sampling functions from Gaussian process posteriors
- Awards: Test of Time and Best Papers
- Marc Deisenroth