2nd November
13:30 | David Forsyth (UIUC) Error analysis, experimental protocols and the replication crisis (video) |
14:00 | Zachary Lipton (CMU) Troubling trends in machine learning scholarship (video) |
14:30 | Workshop and pre-registration overview (video) |
14:45 | Contributed talk (video) An empirical study of the relation between network architecture and complexity (Emir Konuk and Kevin Smith) |
15:00 | Poster session and coffee break |
15:45 | Michela Paganini (FAIR) Preregistration and blind analysis (video) |
16:15 | Contributed talk (video) Learning representational invariance instead of categorization (Alex Hernández and Peter König) |
16:30 | Oguzhan Gencoglu (topdatascience.com) The HARK side of deep learning (video) |
17:00 | Contributed talk (video) Towards generalizable distance estimation by leveraging graph information (Todd Houghton and John Kevin Cava) |
17:15 | Discussion and closing remarks (video) |
Emir Konuk, Kevin Smith | An empirical study of the relation between network architecture and complexity | PDF Bib |
Alex Hernández-García, Peter König | Learning representational invariance instead of categorization | PDF Bib |
John Kevin Cava, Todd Houghton, Hongbin Yu | Towards generalizable distance estimation by leveraging graph information | PDF Bib |
Eimear O' Sullivan, Stefanos Zafeiriou | Extending convolutional pose machines for facial landmark localization in 3D point clouds | PDF Bib |
Mohamed Abbas Hedjazi, Yakup Genç | Learning to inpaint by progressively growing the mask regions | PDF Bib |
João F. Henriques (University of Oxford)
Samuel Albanie (University of Oxford)
Luca Bertinetto (FiveAI)
Jack Valmadre (Google Research)