Subha Maity

Blogs

Mon, 01 Jan 0001 00:00:00 +0000

jhgjhqaj

Contact

Mon, 01 Jan 0001 00:00:00 +0000

Email: ‘‘smaity at uwaterloo dot ca’’

Office: M3-4227, University of Waterloo

CV

Mon, 01 Jan 0001 00:00:00 +0000

Education

PhD in Statistics, University of Michigan, 2024
Master of Statistics, Indian Statistical Institute, 2018
Bachelor of Statistics, Indian Statistical Institute, 2016

Positions

Assistant Professor, Department of Statistics and Actuarial Science, University of Waterloo, 2024-present

Grants

University of Waterloo Startup Grant (2024-2029): CAD 60,000
NSERC Discovery Grant (2026-2032): CAD 185,000 (with a supplement of CAD 12,500)

Publications

Mon, 01 Jan 0001 00:00:00 +0000

Trainees are indicated as (*)

2026+

Chakraborty, A., & Maity, S. (2026). The Statistical Cost of Adaptation in Multi-Source Transfer Learning. (Submitted to JMLR) [preprint]
Xu, M.(*), Maity, S., & Dubin, J. (2026). Robust inference for risk heterogeneity under group imbalance. (To be submitted to Biometrics) [preprint]

2025

Cheng, M.(*), Maity, S., Tian, Q., & Li, P. (2025). Transfer Learning under Group-Label Shift: A Semiparametric Exponential Tilting Approach. (Major revision in Scandinivian Journal of Statistics) [preprint]
Somerstep, S., Polo, F.M., de Oliveira, A.F.M., Mangal, P., Silva, M., Bhardwaj, O., Yurochkin, M. and Maity, S. (2025). Carrot: A cost aware rate optimal router. (Submitted for a review in the Annals of Applied Statistics after major revision) [preprint]
Xu, M.(*), Maity, S., & Dubin, J. (2025). Diagnosis-based mortality prediction for intensive care unit patients via transfer learning. (To be submitted to Canadian Journal of Statistics) [preprint]
Bracale, D., Maity, S., Banerjee, M., & Sun, Y. (2025). Learning the distribution map in reverse causal performative prediction. AISTAT. [paper]
Somerstep, S., Ritov, Y. A., Yurochkin, M., Maity, S., & Sun, Y. (2025). Limitations of refinement methods for weak to strong generalization. COLM. [paper]
Bracale, D., Maity, S., Polo, F. M., Somerstep, S., Banerjee, M., & Sun, Y. (2025). Microfoundation inference for strategic prediction. AISTAT. [paper]

2024

Maity, S., Dutta, D., Terhorst, J., Sun, Y., & Banerjee, M. (2024). A linear adjustment-based approach to posterior drift in transfer learning. Biometrika, 111(1), 31-50. [paper]
Maia Polo, F., Maity, S., Yurochkin, M., Banerjee, M., & Sun, Y. (2024). Weak supervision performance evaluation via partial identification. NeurIPS. [paper]
Maity, S., Agarwal, M., Yurochkin, M., & Sun, Y. (2024, May). An investigation of representation and allocation harms in contrastive learning. ICLR. [paper]
Ngweta, L., Agarwal, M., Maity, S., Gittens, A., Sun, Y., & Yurochkin, M. (2024, November). Aligners: Decoupling llms and alignment. EMNLP. [paper]

2023

Maity, S., Yurochkin, M., Banerjee, M., & Sun, Y. (2022). Understanding new tasks through the lens of training data via exponential tilting. ICLR. [paper]
Maity, S., Mukherjee, D., Banerjee, M., & Sun, Y. (2022). Predictor-corrector algorithms for stochastic optimization under gradual distribution shift. ICLR. [paper]
Ngweta, L., Maity, S., Gittens, A., Sun, Y., & Yurochkin, M. (2023). Simple disentanglement of style and content in visual representations. ICML. [paper]
Bakshi, S., & Maity, S. (2023). Bayes classifier cannot be learned from noisy responses with unknown noise rates. ICLR Tiny Paper. [paper]

2022

Maity, S., Sun, Y., & Banerjee, M. (2022). Minimax optimal approaches to the label shift problem in non-parametric settings. Journal of Machine Learning Research, 23(346), 1-45. [paper]
Maity, S., Sun, Y., & Banerjee, M. (2022). Meta-analysis of heterogeneous data: integrative sparse regression in high-dimensions. Journal of Machine Learning Research, 23(198), 1-50. [paper]
Bhattacharyya, R., Burman, A., Singh, K., Banerjee, S., Maity, S., Auddy, A., Rout, S.K., Lahoti, S., Panda, R. and Baladandayuthapani, V. (2022). Role of multiresolution vulnerability indices in COVID-19 spread in India: a Bayesian model-based analysis. BMJ open, 12(11), p.e056292. [paper]
Kwon, B.C., Kartoun, U., Khurshid, S., Yurochkin, M., Maity, S., Brockman, D.G., Khera, A.V., Ellinor, P.T., Lubitz, S.A. and Ng, K. (2022). RMExplorer: A visual analytics approach to explore the performance and the fairness of disease risk models on population subgroups. In 2022 IEEE Visualization and Visual Analytics (VIS) (pp. 50-54). IEEE. [paper]

2021

Maity, S., Xue, S., Yurochkin, M., & Sun, Y. (2021). Statistical inference for individual fairness. ICLR. [paper]
Maity, S., Mukherjee, D., Yurochkin, M., & Sun, Y. (2021). Does enforcing fairness mitigate biases caused by subpopulation shift?. NeurIPS. [paper]

Research

Mon, 01 Jan 0001 00:00:00 +0000

Current Research Interests

Transfer learning: Transfer learning improves prediction and inference in data-scarce target domains by borrowing information from related source domains. It has emerged as a central topic in machine learning and statistics over the past decade. My research focuses on the development and understanding statistical performance of transfer learning models and methods for a variety of statistical tasks. Selected examples of this work are listed below.

Chakraborty, A., & Maity, S. (2026). The Statistical Cost of Adaptation in Multi-Source Transfer Learning. (Submitted to JMLR) [preprint]

Cheng, M., Maity, S., Tian, Q., & Li, P. (2025). Transfer Learning under Group-Label Shift: A Semiparametric Exponential Tilting Approach. (Major revision in Scandinivian Journal of Statistics) [preprint]

Xu, M., Maity, S., & Dubin, J. (2025). Diagnosis-based mortality prediction for intensive care unit patients via transfer learning. (To be submitted to Canadian Journal of Statistics) [preprint]

Maity, S., Dutta, D., Terhorst, J., Sun, Y., & Banerjee, M. (2024). A linear adjustment-based approach to posterior drift in transfer learning. Biometrika, 111(1), 31-50. [paper]

Maity, S., Yurochkin, M., Banerjee, M., & Sun, Y. (2022). Understanding new tasks through the lens of training data via exponential tilting. ICLR. [paper]

Maity, S., Sun, Y., & Banerjee, M. (2022). Minimax optimal approaches to the label shift problem in non-parametric settings. Journal of Machine Learning Research, 23(346), 1-45. [paper]

Distribution shift: More broadly, I am interested in statistical methods for learning from data that are subject to distribution shift. Beyond transfer learning, my research interest includes areas such as integrative analysis, domain adaptation, performative prediction, etc. Selected works in these areas are listed below.

Teaching

Mon, 01 Jan 0001 00:00:00 +0000

Courses

Winter 2027: STAT 330, Mathematical Statistics
Fall 2026: STAT 845, Statistical Concepts for Data Science
Fall 2025: STAT 441-841/CM 763, Statistical Learning-Classification
Fall 2025: STAT 845, Statistical Concepts for Data Science
Spring 2025: STAT 333, Stochastic Processes I
Winter 2025: STAT 440-840/CM 761, Computational Inference