Andrew Gordon Wilson

UPDATE: I have moved to NYU. If you are a student wanting to work with me, I encourage you to apply to NYU Courant CS, Math, or the Center for Data Science (CDS), and to list me in your application. My new website is https://cims.nyu.edu/~andrewgw. You will be re-directed there shortly.

I lead a machine learning group at Cornell where I advise students in ORIE, Computer Science, Statistics, and CAM. You can check out some of our work here. I also teach classes on Bayesian machine learning and information theory.  I organized the NIPS 2017 symposium on Interpretable Machine Learning. 

I am interested in developing flexible, interpretable, and scalable machine learning models, often involving kernel learning, deep learning, and Gaussian processes.  I am particularly excited about probabilistic approaches. My work has been applied to time series, vision, NLP, spatial statistics, public policy, medicine, and physics.
 
Outside of work, I am a classical pianist who particularly enjoys Glenn Gould's playing of Bach.

News

Upcoming talks: I am giving invited talks at CMStatistics 2017, MSR Cambridge, Cambridge University, UCL Gatsby, Banff International Research Centre (Interface of Statistics and Machine Learning) 2018, DALI 2018, and SIAM ALA (Applied Linear Algebra) 2018, Allerton 2018, and the Toronto Deep Learning Summer School!

Thesis

My thesis provides an introduction to probabilistic non-parametric model construction, Gaussian processes and kernel design, and a vision for scalable and automatic kernel learning, with ideas for future directions.