firstbacksecondback
5 Results
|
Poster
|
Thu 2:30 |
How Do Vision Transformers Work? Namuk Park · Songkuk Kim |
|
|
Spotlight
|
Thu 2:30 |
How Do Vision Transformers Work? Namuk Park · Songkuk Kim |
|
|
Poster
|
Thu 10:30 |
A Loss Curvature Perspective on Training Instabilities of Deep Learning Models Justin Gilmer · Behrooz Ghorbani · Ankush Garg · Sneha Kudugunta · Behnam Neyshabur · David Cardoze · George Dahl · Zachary Nado · Orhan Firat |
|
|
Poster
|
Thu 10:30 |
How many degrees of freedom do we need to train deep networks: a loss landscape perspective Brett Larsen · Stanislav Fort · Nic Becker · Surya Ganguli |
|
|
Poster
|
Mon 2:30 |
The Role of Permutation Invariance in Linear Mode Connectivity of Neural Networks Rahim Entezari · Hanie Sedghi · Olga Saukh · Behnam Neyshabur |