Deep Learning

Winner of TreeHacks ‘23 - Stanford’s annual hackathon. We develop a novel AI-driven system to reduce medical errors in the US healthcare system.

We design a novel data augmentation module for meta learning that creates instance specific augmentations in a few shot setting.

We develop an AI-assisted system which aids a 911 operator to save more lives.

We aim to address the problem of concept-based explainability in natural language understanding tasks.

We show that diffusion models based on a score matching objective are improved with a multi-sample training objective.