Devon Graham

Devon Graham

I study Artificial Intelligence and Machine Learning at the University of British Columbia, where I am working towards a PhD under Professor Kevin Leyton-Brown.

I completed my MSc, also at UBC, under Assistant Professor Siamak Ravanbakhsh. My masters research focused on applying Deep Learning methods to data from structured domains such as graphs or tensors, where the structure of the data helps to inform the learning model.

I have recently started exploring the problem of algorithm configuration, where we seek to set the parameters of a given algorithm so as to achieve good performance across a given set of input instances.

Publications

Procrastinating with Confidence: Near-Optimal, Anytime, Adaptive Algorithm Configuration
Robert Kleinberg, Kevin Leyton-Brown, Brendan Lucier, and Devon Graham. NeurIPS, 2019.
pdf | arXiv | bibtex | poster

Equivariant Entity-Relationship Networks
Devon Graham and Siamak Ravanbakhsh. arXiv, 2019.
pdf | arXiv | bibtex

Deep Models of Interactions Across Sets
Jason Hartford*, Devon Graham*, Kevin Leyton-Brown, and Siamak Ravanbakhsh. ICML, 2018.
pdf | arXiv | bibtex | poster

(* equal contribution)

Teaching

Education

Contact

Department of Computer Science
UNIVERSITY OF BRITISH COLUMBIA
ICICS X568 - 2366 Main Mall
Vancouver, B.C.
V6T 1Z4
Canada
E-mail: drgraham@cs.ubc.ca