COLL-L2: Session 3R: Robustness Reproducibility Replicability |
Session Type: Lecture |
Time: Friday, 8 May, 08:00 - 10:00 |
Location: On-Demand |
Virtual Session: View on Virtual Platform |
Session Chairs: Tulay Adali, University of Maryland, Baltimore County and Peter Schreier, Universität Paderborn, Germany |
COLL-L2.1: THE EMPIRICAL DUALITY GAP OF CONSTRAINED STATISTICAL LEARNING |
Luiz Chamon; University of Pennsylvania |
Santiago Paternain; University of Pennsylvania |
Miguel Calvo-Fullana; University of Pennsylvania |
Alejandro Ribeiro; University of Pennsylvania |
COLL-L2.2: CONTEXT AND UNCERTAINTY MODELING FOR ONLINE SPEAKER CHANGE DETECTION |
Hagai Aronowitz; IBM Research AI |
Weizhong Zhu; IBM Research AI |
COLL-L2.3: MODELING UNCERTAINTY IN PREDICTING EMOTIONAL ATTRIBUTES FROM SPONTANEOUS SPEECH |
Kusha Sridhar; University of Texas at Austin |
Carlos Busso; University of Texas at Austin |
COLL-L2.4: ACCURACY-ROBUSTNESS TRADE-OFF FOR POSITIVELY WEIGHTED NEURAL NETWORKS |
Ana Neacsu; University Politehnica of Bucharest |
Jean-Christophe Pesquet; Université Paris-Saclay, CentraleSupélec, Inria |
Corneliu Burileanu; University Politehnica of Bucharest |
COLL-L2.5: TOWARDS A NEW UNDERSTANDING OF THE TRAINING OF NEURAL NETWORKS WITH MISLABELED TRAINING DATA |
Herbert Gish; Raytheon BBN Technologies |
Jan Silovsky; Apple |
Man-Ling Sung; Chinese University of Hong Kong |
Man-Hung Siu; Apple |
William Hartmann; Raytheon BBN Technologies |
Zhuolin Jiang; Raytheon BBN Technologies |
COLL-L2.6: ON NETWORK SCIENCE AND MUTUAL INFORMATION FOR EXPLAINING DEEP NEURAL NETWORKS |
Brian Davis; Carnegie Mellon University |
Umang Bhatt; Carnegie Mellon University |
Kartikeya Bhardwaj; Carnegie Mellon University |
Radu Marculescu; Carnegie Mellon University |
José Moura; Carnegie Mellon University |