Search Contact information
University of Cambridge Home Department of Engineering
Computational and Biological Learning Lab
University of Cambridge >  Engineering Department >  Information Engineering  >  Computational and Biological Learning Lab  > Carl Edward Rasmussen

Carl Edward Rasmussen

Reader in Information Engineering
Machine Learning Group
Computational and Biological Learning Lab
Department of Engineering
University of Cambridge

Cambridge University
Adjunct Research Scientist
Max Planck Institute for Biological Cybernetics
Tübingen, Germany

Cybernetics logo
Fellow of Darwin College

Research

I have very broad interests in probabilistic inference in machine learning, covering both unsupervised, supervised and reinforcement learning. I'm particularly interested in design and evaluation of non-parametric methods such as Gaussian processes and Dirichlet processes. Exact inference in these models is often intractable, so one needs to resort to approximation methods, such as variational techniques or Markov chain Monte Carlo.

Gaussian Processes

I have co-authored a book with Chris Williams, entitled Gaussian Processes for Machine Learning, MIT Press, 2006, online version. Gaussian processes are a principled, practical, probabilistic approach to learning in kernel machines. The book describes Gaussian process approaches to regression and classification. It also discusses methods for hyperparameter tuning and model selection. Detailed algorithms are given, and demonstrations and a matlab implementation allowing very general covariance structures are available at the book web site. I gave a tutorial lecture on Gaussian processes at the NIPS 2006 conference. I also maintain a web site on Gaussian processes. I have a page discussing the prediction of atmospheric Carbon Dioxide concentrations. Gaussian Processes for Machine Learning cover

Reinforcement Learning

I'm interested in how to speed up reinforcement learning by using a model-based approach with probabilistic models. See a short demo.

Students and Postdocs

David Duvenaud
Roger Frigola
Rowan McAllister
Andrew McHutchon
Mark van der Wilk
Andrew Wilson

Former:

Lehel Csató (postdoc), University of Babes-Bolyai, Romania
Marc Deisenroth, Department of Computing, Imperial College London
Dilan Görür, Yahoo!
Ferenc Huszár, Senior data scientist at PeerIndex
Malte Kuss, Booz Allen Hamilton
Hannes Nickisch, Philips Research Hamburg
Tobias Pfingsten, OC&C Strategy Consultants
Joaquin Quiñonero Candela, Facebook
Yunus Saatçi, Eladian Partners
Ryan Turner, Northrop Grumman

Teaching

1BP7 Part 1B Paper 7, Probability and Statistics
4f13 Machine Learning

Talks

Tutorial on Gaussian Processes at NIPS 2006, slides, pdf.

Publications

My publications can be found at the Machine Learning Group Publications page, or in a limited version at the Department Publications page.

Some worthwhile things on the Web

The book Sustainable Energy - without the hot air, facts about sustainable energy by David MacKay.
What is Science?, by Richard Feynman, 1966.
Inconsistent Maximum Likelihood Estimation: An "Ordinary" Example a simple illustrative example from Radford Neal's blog.

Contact Information

Department of Engineering
Trumpington Street
Cambridge, CB2 1PZ, UK
voice +44 (0) 1223 748 513
fax +44 (0) 1223 332 662
email email address
PGP public key

My office is on the fourth floor of the Baker Building room number BE451.

© Cambridge University Engineering Dept
Information provided by Carl Edward Rasmussen (cer54)
Last updated: August 2nd 2012