dhanya.sridhar <at> mila.quebec
F.04, 6666 Rue St. Urbain
In brief, my research focuses on combining causality and machine learning in service of AI systems that are robust to distribution shifts, adapt to new tasks efficiently, and discover new knowledge alongside us. The topics I work on span causal representation learning to robust supervised prediction. I'm interested in both technical results and practical algorithms that work for data such as text, images, networks, or multiple modalities.
I'm looking for MSc and PhD students to work on research at the intersection of causality and ML. Come join the vibrant intellectual community at Mila. Women and students from underrepresented groups: Mila and my lab are committed to being inclusive spaces -- I strongly encourage you to apply! Read more about applying to work with me.
- Oct 2022: I gave a talk at MSR Montreal on causal machine learning.
- Sep 2022: Our work on Identifiable Deep Generative Models via Sparse Decoding has been accepted to appear in Transactions on Machine Learning Research (TMLR)!
- Aug 2022: I gave a tutorial on causal inference and language at the WASP summer school on the Synthesis of Human Communication in Norrkoeping, Sweden.
Heterogeneous Supervised Topic Models
Dhanya Sridhar, Hal Daumé, D. Blei
Transactions of the Association for Computational Linguistics, 2022
Leveraging Structure Between Environments: Phylogenetic Regularization Incentivizes Disentangled Representations
Elliot I. Layne, Dhanya Sridhar, Jason S. Hartford, M. Blanchette
Identifiable Deep Generative Models via Sparse Decoding
Gemma E. Moran, Dhanya Sridhar, Yixin Wang, D. Blei
Transactions of Machine Learning Research, To Appear, 2022
IFT 6168, Winter 2022
IFT 6168, Winter 2023