Dhanya Sridhar is a Canada CIFAR AI Chair at Mila and an assistant professor at the Department of Computer Science and Operations Research at the University of Montreal.
Sridhar develops theory and machine learning methods to study causal questions. Causal questions are about interventions to the world: how would patients respond if they were given a new therapeutic? What is the effect of a new content moderation policy on social media posts?
Researchers cannot always run experiments to answer such questions — experiments may be costly or unethical. Instead, researchers often collect large-scale datasets about the state of the world: what therapeutics do patients take and what outcomes do we observe? The challenge is that, from non-experimental data, distinguishing between causal relationships and correlations is difficult due to sources of bias such as confounding factors.
Her research focuses on i) theoretically understanding what causal questions can be answered from observed data, and ii) adapting machine learning methods — especially in the context of text and social network data — to estimate causal effects.
- Rising Stars in EECS, 2020
- President's Dissertation-Year Fellowship, UC Santa Cruz, 2017
- Graduate Student Fellowship Honorable Mention, NSF, 2015
Veitch, V., Sridhar, D., & Blei, D. (2020, August). Adapting text embeddings for causal inference. In Conference on Uncertainty in Artificial Intelligence. * Equal contribution
Pryzant, R., Card, D., Jurafsky, D., Veitch, V., & Sridhar, D. (2021, June). Causal Effects of Linguistic Properties. In North American Chapter of the Association for Computational Linguistics: Human Language Technologies.
Sridhar, D., & Getoor, L. (2019, August). Estimating Causal Effects of Tone in Online Debates. In International Joint Conference on Artificial Intelligence.
CIFAR is a registered charitable organization supported by the governments of Canada, Alberta and Quebec, as well as foundations, individuals, corporations and Canadian and international partner organizations.