Past Events
Qualifying ExamDisconnected Manifold Learning for Generative Adversarial Networks |
|
||
Thursday, September 27, 2018, 11:00am |
|||
Real images often lie on a union of disjoint manifolds rather than one globally connected manifold, and this can hinder the training of common Generative Adversarial Networks (GANs). We first show that single generator GANs are unable to correctly model
Speaker: Mahyar Khayatkhoei
Bio
NULL
Location : Hill 482
Committee:
Prof. Ahmed Elgammal (Chair), Prof. Abdeslam Boularias, Prof. Pranjal Awasthi, Prof. William Steiger
Event Type: Qualifying Exam
:
Organization:
Dept. of Computer Science
Department of Computer Science