Tind Technologies (Norway)
Hes-so: ArODES Open Archive (University of Applied Sciences and Arts Western Switzerland / Haute école spécialisée de Suisse occidentale / FH Westschweiz)Not a member yet
15764 research outputs found
Sort by
Wet life
No full textI wake up in the water. Not underwater but just there on the shore. Or on the rocks, by the lake. My body just washed ashore, but I wake up quietly. I try and catch my breath even though I don't need to. Actually, I can breathe underwater. I am wet and warm. I can hear the sound of the water flowing. I feel the waves on my back. I can stay here quietly, or I can get up and play with the dog who is barking. I am in Mads Bycroft's Waterlogue -Four to the Floor. Now that I am completely wet, I feel the water in my body. I feel it in my blood, flowing in my veins. I remember my mother's milk that fed me which transformed into the fluids of my body. I become conscious of the sweat on my skin and the saliva in my mouth. I can feel that I am awake and that my body is fluid. I used to dream in a dry world. It was a nightmare. My society was oppressive, the dry world categorized people using tools like gender, race, sexuality, and determined if our relationships were legal kin, friendship or subordination
Fidex and FidexGlo::from local explanations to global explanations of deep models
No full textDeep connectionist models are characterized by many neurons grouped together in many successive layers. As a result, their data classifications are difficult to understand. We present two novel algorithms which explain the responses of several black-box machine learning models. The first is Fidex, which is local and thus applied to a single sample. The second, called FidexGlo, is global and uses Fidex. Both algorithms generate explanations by means of propositional rules. In our framework, the discriminative boundaries are parallel to the input variables and their location is precisely determined. Fidex is a heuristic algorithm that, at each step, establishes where the best hyperplane is that has increased fidelity the most. The algorithmic complexity of Fidex is proportional to the maximum number of steps, the number of possible hyperplanes, which is finite, and the number of samples. We first used FidexGlo with ensembles and support vector machines (SVMs) to show that its performance on three benchmark problems is competitive in terms of complexity, fidelity and accuracy. The most challenging part was then to apply it to convolutional neural networks. We achieved this with three classification problems based on images. We obtained accurate results and described the characteristics of the rules generated, as well as several examples of explanations illustrated with their corresponding images. To the best of our knowledge, this is one of the few works showing a global rule extraction technique applied to both ensembles, SVMs and deep neural networks
LusTra::waste type prediction through sound and accelerometer data
No full textIn recent years, advancements in waste sorting have been significantly enhanced by the integration of deep learning algorithms. In this regard, LusTra proposes a waste type recognition system using sound and accelerometer data. Two waste types are considered: Polyethylene Terephthalate (PET) bottles and Aluminium cans. Predictions through sound data, converted to Mel spectrograms, with Convolutional Neural Networks (CNN), are promising and result in an accuracy of 89% for PET waste and 90% for aluminium waste. Random forest with features extracted from the accelerometer data provide an accuracy of 76% and 86% for PET and aluminium respectively
Mutations dans le monde du travail ::quelles conséquences pour l'inspection ?
No full textDéveloppement des plateformes, lieux de travail fissurés, télétravail : le monde du travail connaît de profondes mutations. Une étude met en lumière les défis auxquels les inspections du travail sont confrontées à travers le monde ainsi que les stratégies qu’elles ont développées
Enjeux épistémologiques, politiques et méthodologiques de la recherche-action collaborative pour les travailleurs sociaux
No full textPrises de parole ::a conversation
No full textLaurence Rasti photographs to change lives. Her survey of the La Promenade detention facility in La Chaux-de-Fonds serves as a case study for more general questions about the challenges of confinement. It was important to her that those detained be able to play an active part in recording their experiences. Their handling of pinhole boxes transforms the ambient light into photographic traces that are as surprising as they are moving. The harshness of concrete as a barrier to the tradition of prison graffiti challenged the artist to invent an equivalent means of expression, capable of relaying outside the walls the striking testimonies gathered within them. In harnessing the genre of portraiture, to which everyone could lend themselves freely, the exchange of glances, accepted and refused in equal measure, solicits empathy. A contribution by Luca Gnaedinger and an interview with Federica Martini frame this sensitive, conceptually rigorous and powerful artistic work
Pratiquer l'analyse d'image avec les mèmes
No full textCet article restitue le travail pédagogique mené autour des mèmes dans le cours d'Analyse de l'image destiné aux étudiant-e-xs de première année du Bachelor Communication Visuelle
