Meera Krishnamoorthy
About Me
I am an AI Research Engineer at Kira Learning.
I recently completed my PhD in Computer Science at the University of Michigan, where I was advised by Professor Jenna Wiens. My research focused on developing methods to improve the computational efficiency and robustness of clinical machine learning models. Specifically, I addressed
- efficiency with respect to storage and model complexity
- robustness to dataset shift caused by unstable feature-outcome correlations and shifts in censoring patterns
I had the opportunity to apply these methods across a range of clinical tasks, including metagenomic classification, chest X-ray classification, and predicting time to spontaneous labor.
I received my bachelor's degree from Caltech (major: Electrical Engineering, minor: Computer Science), where I developed computational tools for autonomous systems control and astronomy.
Here is my CV.
News
- May 2025: "Cross-Validation for Longitudinal Datasets with Unstable Correlations" was accepted to KDD!
Selected Publications
- Meera Krishnamoorthy, Michael Sjoding, Jenna Wiens.
Cross-Validation for Longitudinal Datasets with Unstable Correlations
Conference Paper, ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD) (2025) - Meera Krishnamoorthy, Jenna Wiens.
Multiple Instance Learning with Absolute Position Information
Conference Paper, AHLI Conference on Health, Inference, and Learning (CHIL) (2024)
[Article] - Meera Krishnamoorthy, Michael W. Sjoding, Jenna Wiens.
Off-label use of artificial intelligence models in healthcare
Comment, Nature Medicine (2024)
[Article] - Meera Krishnamoorthy, Piyush Ranjan, John R. Erb-Downward, Robert P. Dickson, Jenna Wiens.
AMAISE: a machine learning approach to index-free sequence enrichment
Journal Paper, Nature Communications Biology 5, 568 (2022)
[Article] [PDF] [Blog Post] [Video Link]
Contact Information
meera (at) alumni (dot) caltech (dot) edu