SiG-DML-L1: Explainable and Interpretable Learning
Thu, 3 Sep, 10:30 - 12:30
Location: Chamber Music Hall
Session Type: Lecture
Session Chair: Nikos Deligiannis, Vrije Universiteit Brussel
Track: 6: SiG-DML - Signal and Data Analytics for Machine Learning
Thu, 3 Sep, 10:30 - 10:50

SiG-DML-L1.1: Learnable Multi-Window Anisotropic Denoising via Gated Fusion and Residual Refinement
Karen Eguiazarian (Egiazarian), Vladimir Katkovnik, Tampere University, Finland, Finland
Thu, 3 Sep, 10:50 - 11:10

SiG-DML-L1.2: Low Rank properties and architectures in small Image Classification Neural Networks
Mark Sandler, Queen Mary University of London, United Kingdom
Thu, 3 Sep, 11:10 - 11:30

SiG-DML-L1.3: QUANTIFYING EXPLANATION DRIFT IN AUDIO USING POST-HOC EXPLAINABLE CONTINUAL LEARNING
Joan Imbwaga, Manjunath Mulimani, Okko Räsänen, Tampere University, Kenya
Thu, 3 Sep, 11:30 - 11:50

SiG-DML-L1.4: INNER LOOP INFERENCE FOR PRETRAINED TRANSFORMERS: UNLOCKING LATENT CAPABILITIES WITHOUT TRAINING
Jonathan Lys, Vincent Gripon, Bastien Pasdeloup, Axel Marmoret, IMT Atlantique, France; Lukas Mauch, Fabien Cardinaux, Ghouthi Boukli Hacene, Sony Europe, Germany
Thu, 3 Sep, 11:50 - 12:10

SiG-DML-L1.5: SVD-BASED TYPICALITY MAPS FOR OUT-OF-DISTRIBUTION DETECTION IN VISION TRANSFORMERS
Aldo Sean Sartor, Leandro de Souza Rosa, Andriy Enttsel, Mauro Mangia, Riccardo Rovatti, University of Bologna, Italy
Thu, 3 Sep, 12:10 - 12:30

SiG-DML-L1.6: COMBINING CONVOLUTION AND DELAY LEARNING IN RECURRENT SPIKING NEURAL NETWORKS
Lúcio Folly Sanches Zebendo, Eleonora Cicciarella, Michele Rossi, University of Padova, Brazil