Dougal J. Sutherland
I'm a fifth-year Ph.D. student in the CS department
at Carnegie Mellon University,
working with Jeff Schneider on machine learning.
My work is supported by a Sandia Campus Executive Program fellowship — thanks, Sandia!
I'm primarily interested in kernel-type methods for learning on sets and distributions
and in active learning, particularly in nonstandard settings such as matrices and searching for large-scale patterns.
research code on github,
Below, ** denotes equal contribution.
Jeff Schneider (chair),
Proposal: Scalable, Flexible, and Active Learning on Distributions.
Journal and Peer-Reviewed Conference Papers
Dynamical Mass Measurements of Contaminated Galaxy Clusters Using Machine Learning.
Accepted to The Astrophysical Journal (ApJ).
Linear-time Learning on Distributions with Approximate Kernel Embeddings.
On the Error of Random Fourier Features.
A Machine Learning Approach for Dynamical Mass Measurements of Galaxy Clusters.
The Astrophysical Journal (ApJ), 803, 50 (2015).
Active learning and search on low-rank matrices.
Nonparametric kernel estimators for image classification.
Managing User Requests with the Grand Unified Task System (GUTS).
Finding Representative Objects with Sparse Modeling.
CMU 10-725 Optimization course project. (Best poster award.)
Kernels on Sample Sets via Nonparametric Divergence Estimates.
Grounding Conceptual Knowledge with Spatio-Temporal Multi-Dimensional Relational Framework Trees.
University of Oklahoma Artificial Intelligence and Robotics Technical Report #1138 (2012).
Integrating Human Knowledge into a Relational Learning System.
Swarthmore College B.A. thesis.