Charles Lu, PhD student at MIT Media Lab

Charles Lu

PhD Researcher at MIT

an email adress

About Me

I am a PhD student in the Camera Culture group at the MIT Media Laboratory. My research focuses on data-centric machine learning. I am supported by the National Science Foundation's Graduate Research Fellowship.

I study how data influences machine learning models through data attribution and incentive alignment frameworks.

Outside the lab, I enjoy activites:

Synthesizers
Pottery

Research Focus

My current research focus

Experience

2025

Research Intern Microsoft Research, Redmond

Working on training data attribution for large language models in the Deep learning Group.

2024

Visiting Researcher University of Chicago

Explored data markets and data escrow mechanisms in the Department of Computer Science at the University of Chicago.

2020–2022

Machine Learning Researcher Harvard Medical School

Developed fairness and uncertainty quantification methods for medical imaging at the Martinos Center.

2019–2022

Machine Learning Scientist Massachusetts General Hospital

Built deep learning models for medical applications such as stroke detection and aortic aneurysm quantification.

2018

Research Intern Oak Ridge National Laboratory

Applied machine learning to molecular dynamics simulations on supercomputing infrastructure.

2018

Research Intern NASA

Analyzed hurricane damage using satellite imagery for environmental policy planning.

Selected papers

Data Acquisition via Experimental Design for Decentralized Data Markets

NeurIPS 2024

Developed a framework for optimizing budget-aware data acquisition.

Federated Conformal Predictors for Distributed Uncertainty Quantification

ICML 2023

Introduced method for uncertainty quantification in federated learning scenarios.

Improving Trustworthiness in Automatic Disease Severity Rating with Ordinal Conformal Prediction

MICCAI 2022

Developed ordinal conformal prediction for disease severity assessment in medical imaging.

Fair Conformal Predictors for Applications in Medical Imaging

AAAI 2022

Proposed fairness in uncertainty quantification across diverse patient populations to address bias in medical AI systems.

Quantification of the Thoracic Aorta and Detection of Aneurysm at CT: Validation of a Fully Automatic Methodology

Radiology: Artificial Intelligence

Designed and validated automated system for aortic aneurysm detection in CT scans.