HAL Portal UPPA (University of Pau and the Pays de l'Adour)
Not a member yet
42255 research outputs found
Sort by
Biocompatibility and in situ colonization of concrete and mortar using crushed seashell waste as a complete aggregate replacement (sand and gravel)
No full textThe Traveller's Body in the Literature, Civilisation and Cultures of the English-Speaking World
No full textInternational audienceThis collection of articles follows on from the research work regarding the body—the face, the reader’s body, the artist’s body—and is among the topics addressed by one of the teams of the ALTER laboratory at Université de Pau et des Pays de l’Adour. These topics center around the representations of the individual and their relationship with the world as they move and migrate. The contributors study different forms of embodiment and corporeality as experienced by travellers who have chosen to travel. They show how the process of discovery and the confrontation with the unknown lead the traveller’s body to be at odds with the Other and they also illustrate the physicality of the journey, rooted in its material nature, by examining the objects used and the clothes worn by the travellers. These appendages and extensions have a paradoxical function and turn out to be both impediments and sources of empowerment which come to define the self that carries them on their journey. This work also sheds light on individuals on the move, evolving among foreign landscapes: the interactions between their bodies and these new and often impressive surroundings are such a source of liberation that they generate a form of awakening and a reconstruction of sometimes fragmented identities. This book results from the collaboration between academics working on travel and the body, and students of Université de Pau et des pays de l'Adour who proposed their own artistic interpretation of the themes and issues connected with the traveller's body
EERO: Early Exit with Reject Option for Efficient Classification with limited budget
No full textThe increasing complexity of advanced machine learning models requires innovative approaches to manage computational resources effectively. One such method is the Early Exit strategy, which allows for adaptive computation by providing a mechanism to shorten the processing path for simpler data instances. In this paper, we propose EERO, a new methodology to translate the problem of early exiting to a problem of using multiple classifiers with reject option in order to better select the exiting head for each instance. We calibrate the probabilities of exiting at the different heads using aggregation with exponential weights to guarantee a fixed budget .We consider factors such as Bayesian risk, budget constraints, and head-specific budget consumption. Experimental results, conducted using a ResNet-18 model and a ConvNext architecture on Cifar and ImageNet datasets, demonstrate that our method not only effectively manages budget allocation but also enhances accuracy in overthinking scenarios
A novel DFT for associating fluids confined in nanopores and its application to water
No full textInternational audienc
Présentation de l’ouvrage de Pilar Gonzalez Bernaldo, Argentina hasta la muerte. Policas de nacionalidad y practicas de naturalizacion, siglos XIX-XX, FCE, 2024
No full textInternational audienc
Global Positioning System‐Derived Metrics and Machine Learning Models for Injury Prediction in Professional Rugby Union Players
No full textInternational audienceIn sports, injury prevention is a key factor for success. Although injuries are challenging to predict, new technologies and the application of data science can provide valuable insights. This study aimed to predict injury risk among professional rugby union players using machine learning (ML) models. We analyzed data from 63 professional rugby union players during three seasons, categorized them into forwards and backs, and further classified them into five specific positions (tight five, back row, scrum‐half, inside backs, outside backs). The dataset included GPS data and derived metrics such as total workload in the 1, 2, and 3 weeks prior to injury, acute‐to‐chronic workload ratio over different time windows, monotony, and strain. Injury prediction was assessed separately for different player positions using five ML classification models: logistic regression, naïve Bayes (NB), support vector machine, random forest (RF), and eXtreme gradient boosting (XGBoost). RF performed best for forwards overall, with XGBoost excelling in the tight five and SVM in the back row, whereas among backs, RF led for inside backs and NB for outside backs. Additionally, feature importance plots were used to examine the impact of various factors on injury occurrence. In conclusion, our ML‐based approach can effectively predict injuries, with average F1 scores up to 0.66 (± 0.14), particularly when applying a combination of GPS‐derived metrics. Additionally, key characteristics indicative of injury for players in various positions have been successfully identified. These findings underscored the potential of ML to enhance injury prediction and inform tailored training strategies for athletes
