Video: Salma Tarmoun, "Gradient Descent and Attention Models: Challenges Posed by the Softmax Function"

Video From 24w5297: Mathematics of Deep Learning

Salma Tarmoun, University of Pennsylvania

Tuesday, June 11, 2024 16:30 - 17:03

Gradient Descent and Attention Models: Challenges Posed by the Softmax Function

Salma Tarmoun, Video: Gradient Descent and Attention Models: Challenges Posed by the Softmax Function

Salma Tarmoun, Gradient Descent and Attention Models: Challenges Posed by the Softmax Function, Mathematics of Deep Learning, BIRS, BIRS talk, 24w5297, math, mathematics, video

Banff International Research Station

presentation

mpeg4

38M

http://www.birs.ca/events/2024/5-day-workshops/24w5297/videos

http://www.birs.ca/events/2024/5-day-workshops/24w5297/videos/watch/202406111630-Tarmoun.html

Download this video (38M)