About
Appointed Canada CIFAR AI Chair – 2021
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.
In brief, Sridhar’s research focuses on using causal models to develop large-scale AI systems that are robust to unseen contexts, adapt to new tasks efficiently, and guide the process of scientific discovery. Her group’s research spans causal representation learning, robust prediction especially in large autoregressive models, interpretability, and causal discovery.
Awards
- Rising Stars in EECS, 2020
- President's Dissertation-Year Fellowship, UC Santa Cruz, 2017
Relevant Publications
- Mittal, S., Elmoznino, E., Gagnon, L., Bhardwaj, S., Sridhar, D., & Lajoie, G. (2024). Does learning the right latent variables necessarily improve in-context learning? [Preprint]. arXiv.
- Montagna, F., Cairney-Leeming, M., Sridhar, D., & Locatello, F. (2024). Demystifying amortized causal discovery with transformers [Preprint]. arXiv.
- Kasetty, T., Mahajan, D., Dziugaite, G. K., Drouin, A., & Sridhar, D. (2024). Evaluating interventional reasoning capabilities of large language models [Preprint]. arXiv.
- Feder, A., Keith, K. A., Manzoor, E., Pryzant, R., Sridhar, D., Wood-Doughty, Z., Eisenstein, J., Grimmer, J., Reichart, R., Roberts, M. E., Stewart, B. M., Veitch, V., & Yang, D. (2022). Causal inference in natural language processing: Estimation, prediction, interpretation and beyond. Transactions of the Association for Computational Linguistics, 10, 1138-1158.
- Moran, G.E., Sridhar, D., Wang, Y. and Blei, D. (2021) Identifiable Deep Generative Models via Sparse Decoding. Transactions on Machine Learning Research.