Postdoctoral Researcher

The petitioner proposes to continue his research on effectively mitigating errors in machine learning and statistical frameworks. This work focuses on improving the reliability and accuracy of computational models used to address pressing national issues in data science and automated decision-making.

The petitioner proposes to develop scalable and data-efficient AI systems that can be fine-tuned for critical sectors like healthcare, education, and manufacturing. The work focuses on optimizing performance in data-sparse domains and designing task-specific synthetic data strategies for large language models and vision-language models.

The petitioner proposes to develop machine learning methods and mathematical models with high accuracy to enhance decision-making and optimize system efficiency. The work focuses on optimal-sample strategies for multi-armed bandit problems, generalization guarantees via stability analysis, and fair data-sharing mechanisms for heterogeneous user groups.

The petitioner proposes to advance modeling techniques that blend machine learning with statistical analysis to strengthen predictive analytics and improve decision-making across manufacturing, logistics, and healthcare. Her work focuses on developing scalable approaches that can handle complex, data-rich environments to achieve higher efficiency and reliable system performance. The endeavor aims to provide incremental gains that translate into large economic and public benefits for U.S. infrastructure.