TA4b: Theory of Deep Learning (Invited) |
| Session Type: Oral |
| Time: Tuesday, November 5, 10:15 - 11:55 |
| Location: Heather |
| Session Chairs: Richard Baraniuk, Rice University and Santiago Segarra, Rice University |
| TA4b-1: A SPLINE THEORY OF DEEP NETWORKS |
| Randall Balestriero; Rice University |
| Richard Baraniuk; Rice University |
| TA4b-2: INFORMATION IN THE WEIGHTS AND EMERGENT PROPERTIES OF TRAINING DEEP NEURAL NETWORKS |
| Alessandro Achille; University of California, Los Angeles |
| Stefano Soatto; University of California, Los Angeles |
| TA4b-3: UNDERSTANDING GENERALIZATION IN NEURAL NETS VIA VISUALIZATION |
| W Huang; University of Maryland |
| Zeyad Emam; University of Maryland |
| Micah Goldblum; University of Maryland |
| Liam Fowl; University of Maryland |
| Justin Terry; University of Maryland |
| Tom Goldstein; University of Maryland |
| TA4b-4: ON EXACT COMPUTATION WITH AN INFINITELY WIDE NEURAL NET |
| Sanjeev Arora; Princeton University |
| Simon S. Du; Carnegie Mellon University |
| Wei Hu; Princeton University |
| Zhiyuan Li; Princeton University |
| Ruslan Salakhutdinov; Carnegie Mellon University |
| Ruosong Wang; Carnegie Mellon University |
| TA4b-5: DO IMAGENET CLASSIFIERS GENERALIZE TO IMAGENET? |
| Benjamin Recht; University of California, Berkeley |
| Rebecca Roelofs; University of California, Berkeley |
| Ludwig Schmidt; University of California, Berkeley |
| Vaishaal Shankar; University of California, Berkeley |