Zobrazeno 1 - 10
of 569
pro vyhledávání: '"A. Soutif"'
Recent research identified a temporary performance drop on previously learned tasks when transitioning to a new one. This drop is called the stability gap and has great consequences for continual learning: it complicates the direct employment of cont
Externí odkaz:
http://arxiv.org/abs/2406.05114
Autor:
Goswami, Dipam, Soutif--Cormerais, Albin, Liu, Yuyang, Kamath, Sandesh, Twardowski, Bartłomiej, van de Weijer, Joost
Continual learning methods are known to suffer from catastrophic forgetting, a phenomenon that is particularly hard to counter for methods that do not store exemplars of previous tasks. Therefore, to reduce potential drift in the feature extractor, e
Externí odkaz:
http://arxiv.org/abs/2405.19074
Broad, open source availability of large pretrained foundation models on the internet through platforms such as HuggingFace has taken the world of practical deep learning by storm. A classical pipeline for neural network training now typically consis
Externí odkaz:
http://arxiv.org/abs/2405.18069
Autor:
Magistri, Simone, Trinci, Tomaso, Soutif-Cormerais, Albin, van de Weijer, Joost, Bagdanov, Andrew D.
Exemplar-Free Class Incremental Learning (EFCIL) aims to learn from a sequence of tasks without having access to previous task data. In this paper, we consider the challenging Cold Start scenario in which insufficient data is available in the first t
Externí odkaz:
http://arxiv.org/abs/2402.03917
Autor:
Soutif--Cormerais, Albin, Carta, Antonio, Cossu, Andrea, Hurtado, Julio, Hemati, Hamed, Lomonaco, Vincenzo, Van de Weijer, Joost
Online continual learning aims to get closer to a live learning experience by learning directly on a stream of data with temporally shifting distribution and by storing a minimum amount of data from that stream. In this empirical evaluation, we evalu
Externí odkaz:
http://arxiv.org/abs/2308.10328
Neural networks are very effective when trained on large datasets for a large number of iterations. However, when they are trained on non-stationary streams of data and in an online fashion, their performance is reduced (1) by the online setup, which
Externí odkaz:
http://arxiv.org/abs/2306.16817
Publikováno v:
Principia: An International Journal of Epistemology, Vol 28, Iss 1 (2024)
Using semantic and syntactic methods, we prove the compatibility of the truth of universally quantified slurring sentences of the form [all Ss are Ns] or [all Ss are S*s] with the existential core of moral and semantic innocence. We also show that pr
Externí odkaz:
https://doaj.org/article/ef1640a3a6704dd9a83393031c0ec081
In class-incremental learning, an agent with limited resources needs to learn a sequence of classification tasks, forming an ever growing classification problem, with the constraint of not being able to access data from previous tasks. The main diffe
Externí odkaz:
http://arxiv.org/abs/2106.11930
Autor:
L. Strohmenger, E. Sauquet, C. Bernard, J. Bonneau, F. Branger, A. Bresson, P. Brigode, R. Buzier, O. Delaigue, A. Devers, G. Evin, M. Fournier, S.-C. Hsu, S. Lanini, A. de Lavenne, T. Lemaitre-Basset, C. Magand, G. Mendoza Guimarães, M. Mentha, S. Munier, C. Perrin, T. Podechard, L. Rouchy, M. Sadki, M. Soutif-Bellenger, F. Tilmant, Y. Tramblay, A.-L. Véron, J.-P. Vidal, G. Thirel
Publikováno v:
Hydrology and Earth System Sciences, Vol 27, Pp 3375-3391 (2023)
Large datasets of long-term streamflow measurements are widely used to infer and model hydrological processes. However, streamflow measurements may suffer from what users can consider anomalies, i.e. non-natural records that may be erroneous streamfl
Externí odkaz:
https://doaj.org/article/e97086dbc2be47ed969dc6a433972466
The goal of this paper is to simulate the voters behaviour given a voting method. Our approach uses a multi-agent simulation in order to model a voting process through many iterations, so that the voters can vote by taking into account the results of
Externí odkaz:
http://arxiv.org/abs/2101.11538