Danica J. Sutherland

Canada CIFAR AI Chair

Academic Affiliations

Assistant Professor – University of British Columbia (Computer Science); Member - Centre for Artificial Intelligence Decision-Making and Action (University of British Columbia); Member - Pacific Interdisciplinary Hub on Optimal Transport

Areas of Expertise

Machine learning; deep learning; kernel methods; generative models; two-sample testing; representation learning; statistical learning theory

Danica seeks to make improvements in representation learning – a core tool of modern machine learning.

Going deeper in representation learning

Danica Sutherland seeks to make improvements in representation learning – a core tool of modern machine learning – especially for problems with complex structure, or where a variety of problems must be worked on concurrently. A major goal of her research is to expand options in the architecture and training of deep models, especially via kernel methods, to more easily learn powerful representations in a variety of settings. A major component of this work is on better methods to understand differences between distributions. This can take the form of distinguishing two fixed datasets (as in two-sample testing), as in telling whether treatment or control groups differ. These methods are also key to helping generative models (such as GANs) match their target distributions. Danica also works on related problems in finding good representations of datasets without “strong” labels (unsupervised or weakly-supervised learning), and in finding models that can generalize well to environments different than the ones where they were trained.

Danica is a Canada CIFAR AI Chair at Amii and an Assistant Professor in the Computer Science Department at the University of British Columbia. Her papers have appeared at venues such as the Neural Information and Processing Systems conference (NeurIPS), the International Conference on Machine Learning (ICML), the Artificial Intelligence and Statistics conference (AISTATS), and the International Conference on Learning Representations (ICLR). She has served as an Area Chair or Senior Program Committee member at NeurIPS, AISTATS, and AAAI, jointly given a tutorial session at NeurIPS, given invited talks to venues including a large ICML workshop and the International Congress on Industrial and Applied Mathematics, and taught at École polytechnique’s Data Science Summer School. Previously, Danica held positions at the Toyota Technological Institute at Chicago and the Gatsby Computational Neuroscience Unit at University College London; she earned her Ph.D.

Featured Articles

Danica’s papers have appeared at conferences such as NeurIPS, ICML, AISTATS, and ICLR.

Connect with the community

Get involved in Alberta's growing AI ecosystem! Speaker, sponsorship, and letter of support requests welcome.

Explore training and advanced education

Curious about study options under one of our researchers? Want more information on training opportunities?

Harness the potential of artificial intelligence

Let us know about your goals and challenges for AI adoption in your business. Our Investments & Partnerships team will be in touch shortly!