SiG-DML_L1: Learning Theory and Algorithms |
| Session Type: Lecture |
| Time: Tuesday, August 30, 11:20 - 13:00 |
| Location: Atlantic 3 |
| Session Chair: Emilie Chouzenoux, Inria Saclay |
| SiG-DML_L1.1: A LOCAL MIXUP TO PREVENT MANIFOLD INTRUSION |
| Raphael Baena, Lucas Drumetz, Vincent Gripon, IMT Atlantique, France |
| SiG-DML_L1.2: HIGH-DIMENSIONAL DATA LEARNING BASED ON TENSORIAL-SINGULAR SPACE OF TENSOR TRAIN CORES |
| Ouafae Karmouda, Jeremie Boulanger, Remy Boyer, University of Lille, France |
| SiG-DML_L1.3: TRAINABLE COMPOUND ACTIVATION FUNCTIONS FOR MACHINE LEARNING |
| Paul Baggenstoss, Fraunhofer, Germany |
| SiG-DML_L1.4: AN INVARIANT MATCHING PROPERTY FOR DISTRIBUTION GENERALIZATION UNDER INTERVENED RESPONSE |
| Kang Du, Yu Xiang, University of Utah, United States |
| SiG-DML_L1.5: A DIFFERENTIABLE SHORT-TIME FOURIER TRANSFORM WITH RESPECT TO THE WINDOW LENGTH. |
| Maxime Leiber, INRIA, France; Axel Barrau, Offroad, France; Yosra Marnissi, Dany Abboud, Safran Tech, France |