IID Relaxation by Logical Expressivity: A Research Agenda for Fitting Logics to Neurosymbolic Requirements

Autor: Stol, Maarten C., Mileo, Alessandra

Neurosymbolic background knowledge and the expressivity required of its logic can break Machine Learning assumptions about data Independence and Identical Distribution. In this position paper we propose to analyze IID relaxation in a hierarchy of log

Externí odkaz: http://arxiv.org/abs/2404.19485

Self-supervised Video Representation Learning with Cross-Stream Prototypical Contrasting

Autor: Toering, Martine, Gatopoulos, Ioannis, Stol, Maarten, Hu, Vincent Tao

Instance-level contrastive learning techniques, which rely on data augmentation and a contrastive loss function, have found great success in the domain of visual representation learning. They are not suitable for exploiting the rich dynamical structu

Externí odkaz: http://arxiv.org/abs/2106.10137

FlipOut: Uncovering Redundant Weights via Sign Flipping

Autor: Apostol, Andrei, Stol, Maarten, Forré, Patrick

Modern neural networks, although achieving state-of-the-art results on many tasks, tend to have a large number of parameters, which increases training time and resource usage. This problem can be alleviated by pruning. Existing methods, however, ofte

Externí odkaz: http://arxiv.org/abs/2009.02594

Pruning via Iterative Ranking of Sensitivity Statistics

Autor: Verdenius, Stijn, Stol, Maarten, Forré, Patrick

With the introduction of SNIP [arXiv:1810.02340v2], it has been demonstrated that modern neural networks can effectively be pruned before training. Yet, its sensitivity criterion has since been criticized for not propagating training signal properly

Externí odkaz: http://arxiv.org/abs/2006.00896

Mixing Consistent Deep Clustering

Autor: Lutscher, Daniel, Hassouni, Ali el, Stol, Maarten, Hoogendoorn, Mark, Nicosia, Giuseppe, Ojha, Varun, La Malfa, Emanuele, La Malfa, Gabriele, Jansen, Giorgio, Pardalos, Panos M., Giuffrida, Giovanni, Umeton, Renato

Publikováno v: Lutscher, D, Hassouni, A E, Stol, M & Hoogendoorn, M 2022, Mixing Consistent Deep Clustering . in G Nicosia, V Ojha, E La Malfa, G La Malfa, G Jansen, P M Pardalos, G Giuffrida & R Umeton (eds), Machine Learning, Optimization, and Data Science : [Proceedings] 7th International Conference, LOD 2021, Grasmere, UK, October 4–8, 2021, Revised Selected Papers, Part I . vol. 1, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13163 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 124-137, 7th International Conference on Machine Learning, Optimization, and Data Science, LOD 2021, Virtual, Online, 4/10/21 . https://doi.org/10.1007/978-3-030-95467-3_10
Machine Learning, Optimization, and Data Science: [Proceedings] 7th International Conference, LOD 2021, Grasmere, UK, October 4–8, 2021, Revised Selected Papers, Part I, 1, 124-137
Machine Learning, Optimization, and Data Science ISBN: 9783030954666

Finding well-defined clusters in data represents a fundamental challenge for many data-driven applications, and largely depends on good data representation. Drawing on literature regarding representation learning, studies suggest that one key charact

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::83481d39df910b30358f926612eb1e2a
https://doi.org/10.1007/978-3-030-95467-3_10