Uploaded September 2024 | Updated September 2026, 2 weeks ago
With the 2024 Paralympics underway in Paris, we talked to a 2028 Paralympic hopeful at NYU—Tisch sophomore Katie Kubiak, who qualified for and will be competing with the US National Paralympic Swimming team beginning in 2025. While a full-time student in film & TV, she’s also been training at Palladium and setting American Paralympic records in freestyle and breaststroke. #VioletPride
With the 2024 Paralympics underway in Paris, we talked to a 2028 Paralympic hopeful at NYU—Tisch sophomore Katie Kubiak, who qualified for and will be competing with the US National Paralympic Swimming team beginning in 2025. While a full-time student in film & TV, she’s also been training at Palladium and setting American Paralympic records in freestyle and breaststroke. #VioletPride






![How Naturalistic Learning Algorithm PooDLe Works
Most AI systems trained on carefully curated datasets struggle when faced with the messy reality of naturalistic video. PooDLe [https://arxiv.org/abs/2408.11208], a newly published work, addresses this problem.
“We were interested in building embodied learning algorithms that could learn directly from video streams. Says Mengye Ren, author of the paper and assistant professor of computer science and data science at NYU.
The algorithms name comes from a combination of a pooled loss function and a dense loss function used in PooDLes architecture.
Visit nyu.edu/news for updates on this story and more. How Naturalistic Learning Algorithm PooDLe Works](https://i.ytimg.com/vi/uUotGDIibu8/mqdefault.jpg)



