|This is an old website.
I have now moved to CMU.
Please visit: http://www.cs.cmu.edu/~andrewgw
You will be redirected shortly.
Andrew Gordon Wilson
Postdoctoral Research Fellow in the Sailing Lab at Carnegie Mellon University
[PhD Thesis] [CV] [Papers] [Talks]
I have broad interests in machine learning and statistics. I am particularly interested in developing kernel methods, Gaussian processes, and Bayesian nonparametric models, for scalable automatic pattern discovery and extrapolation, and representation learning. Much of this work involves automatic and expressive approaches to kernel learning, versus hand crafting of features.
In January 2014 I completed my PhD dissertation, "Covariance Kernels for Fast Automatic Pattern Discovery and Extrapolation with Gaussian processes" (news story), in the Machine Learning Group at the University of Cambridge, where I am a member of Trinity College.
Outside of work, I am a classical pianist who particularly enjoys Glenn Gould's playing of Bach.
I am also interested in modern physics, I write essays and fiction, and I enjoy squash, tennis, badminton, soccer, and chess.
- Our paper, "Fast Kernel Learning for Multidimensional Pattern Extrapolation" has been accepted to NIPS 2014! [PDF, BibTeX]
- I am co-organising the NIPS 2014 Workshop
"Modern Nonparametrics 3: Automating the Learning Pipeline"!
- I gave a lecture series on "Kernel Methods for Representation Learning" at MLSS 2014 Pittsburgh!
[Lecture 1+2, Lecture 3+4]
- My PhD Thesis:
Covariance Kernels for Fast Automatic Pattern Discovery and Extrapolation with Gaussian Processes
- Video Lecture on Spectral Mixture (SM) Kernels
A note on GPRNs and changepoints