MIT CSAIL Alliances Student Spotlight | Sharut Gupta
Автор: MIT CSAIL Alliances
Загружено: 2025-05-19
Просмотров: 283
How can AI models learn from the world around us without being told what to look for?
Sharut Gupta, a third-year PhD student at MIT CSAIL, is exploring that question through contrastive learning. Working with Associate Professor Stefanie Jegelka, Gupta is building AI systems that learn rich, meaningful representations from unlabeled data. Her research examines how these models can be sensitive to the right features for a given task—whether it’s clinical decision-making or autonomous driving—while remaining fair and adaptable. Gupta also shares how self-supervised learning can reduce bias and improve generalization, paving the way for AI tools that scale reliably across industries.
For Sharut, truly useful AI should make life easier, augmenting human decisions without needing constant retraining.
Learn more about Sharut’s research here: https://www.mit.edu/~sharut/
Read Sharut’s full Student Spotlight here: https://cap.csail.mit.edu/engage/spot...
Connect with CSAIL Alliances:
On our site: https://cap.csail.mit.edu/about-us/me...
On X: https://x.com/csail_alliances
On LinkedIn: / mit-csail
Доступные форматы для скачивания:
Скачать видео mp4
-
Информация по загрузке:
