643 research outputs found
Proxenoi e asylia nei Symbola di Gauthier
No full textIn the framework of a re-examination of Symbola, the volume dedicated by Gauthier to foreigners and justice, the author of this essay focuses his attention on chapter I, in particular on the figure and role of the proxenos (taking into account the main opinions subsequently expressed in literature), and on chapter V, in particular on the role of protection of the foreigner ensured by the grants of asylia. To clarify the scope of this privilege, it was deemed useful to compare the traditional way of understanding the sylai, i.e. the ‘saisies’ in Gauthier’s language, with the critique of this opinion carried out by B. Bravo in 1980. At the end of his analysis, the author holds the view that many of the research directions on these topics indicated by Gauthier, deserve to be taken into serious consideration by those who wish to continue the study of relations between citizens and foreigners in the Greek world.Nel quadro di un riesame di Symbola, il volume dedicato da Gauthier agli stranieri e la giustizia, l’autore del presente saggio concentra la sua attenzione, da un lato, sul capitolo I, in particolare sulla figura e sul ruolo del proxenos (tenendo conto delle principali opinioni espresse successivamente in dottrina), dall’altro, sul capitolo V, in particolare sul ruolo di protezione dello straniero assicurato dalle concessioni di asylia. Per chiarire la portata di tale privilegio si è ritenuto utile mettere a confronto il modo tradizionale di intendere le sylai, ossia le ‘saisies’ nel linguaggio di Gauthier, con la critica che di tale opinione ha svolto B. Bravo nel 1980. Si è potuto così constatare che molte delle direzioni di ricerca su questi argomenti, indicate da Gauthier, meritano di essere prese in seria considerazione da chi voglia approfondire in generale i rapporti fra cittadini e stranieri nel mondo greco
Performative prediction : expanding theoretical horizons
No full textCette thèse aborde certaines des limitations du cadre de la prédiction performative, qui consiste à apprendre des modèles influençant les données qu’ils sont censés prédire. Je propose des solutions pour repousser les limites de ce cadre, en explorant et en identifiant de nouveaux domaines où son application peut être étendue. La thèse est structurée en trois chapitres, comme décrit ci-après.
Le Chapitre 1 offre un aperçu complet du cadre de la prédiction performative, y compris une vue d’ensemble détaillée de la notation préliminaire (Section 1.1) et des concepts nécessaires à la compréhension du cadre, y compris les concepts de solution (Section 1.2) et l’algorithme de Minimisation de Risque Répété (Section 1.3). La notation de ce chapitre est tirée de l’article original sur la prédiction performative afin de garantir une compréhension fondamentale. De plus, la Section 1.4 introduit la relation entre la prédiction performative et les inégalités variationnelles, qui seront abordées plus en détail au Chapitre 3.
Le Chapitre 2 présente la contribution principale de cette thèse, en analysant le cadre de la prédiction performative en présence de réseaux neuronaux avec une fonction de perte non convexe. L’accent est mis sur la recherche de classificateurs performativement stables, c’est-à-dire optimaux pour la distribution de données qu’ils induisent. Ce chapitre introduit de nouvelles hypothèses et des garanties de convergence significativement plus fortes pour la méthode RRM (Section 2.3). Ces garanties sont les premières à démontrer l’applicabilité de RRM aux réseaux neuronaux, qui sont difficiles à analyser en raison de leur non-convexité. En guise d’illustration, nous introduisons une procédure de rééchantillonnage qui modélise des changements de distribution réalistes et montrons qu’elle satisfait nos hypothèses (Section 2.4). Nous étayons notre théorie en montrant qu’il est possible d’apprendre des classificateurs performativement stables avec des réseaux neuronaux faisant des prédictions sur des données réelles qui changent selon notre procédure proposée (Section 2.5). Ce travail représente une étape cruciale pour combler le fossé entre la prédiction performative théorique et les applications pratiques.
Le Chapitre 3 conclut la thèse en résumant les principales conclusions et contributions et en esquissant de futures directions de recherches. Notamment, il explore l’utilisation des inégalités variationnelles pour aborder et surmonter une limitation significative des travaux antérieurs qui régissent la force des effets performatifs. Cette recherche vise à étendre l’analyse à des scénarios avec des effets performatifs importants et à élargir l’applicabilité du cadre, ouvrant la voie à des solutions plus complètes dans la prédiction performative.This thesis addresses some of the limitations in the framework of performative prediction, which involves learning models that influence the data they intend to predict. I provide solutions to push the boundaries of this framework, exploring and identifying new domains where its application can be extended. The thesis is structured into three chapters, as described in the following.
Chapter 1 offers a comprehensive background on the framework of performative prediction, including a detailed overview of the preliminary notation (Section 1.1) and concepts necessary for understanding the framework, including the solution concepts (Section 1.2) and the Repeated Risk Minimization algorithm (Section 1.3). The notation in this chapter is sourced from the original performative prediction paper to ensure a foundational understanding. Furthermore, Section 1.4 introduces the relationship between performative prediction and variational inequalities, which will be further discussed in Chapter 3.
Chapter 2 introduces the main contribution of this thesis, analyzing the performative prediction framework in the presence of neural networks with non-convex loss functions. The focus is on finding classifiers that are performatively stable, meaning they are optimal for the data distribution they induce. This chapter introduces new assumptions and significantly stronger convergence guarantees for the RRM method (Section 2.3). These guarantees are the first to demonstrate the applicability of RRM to neural networks, which are challenging to analyze due to their non-convexity. As an illustration, we introduce a resampling procedure that models realistic distribution shifts and show that it satisfies our assumptions (Section 2.4). We support our theory by showing that one can learn performative stable classifiers with neural networks making predictions about real data that shift according to our proposed procedure (Section 2.5). This work represents a critical step towards bridging the gap between theoretical performative prediction and practical applications.
Chapter 3 concludes the thesis by summarizing the key findings and contributions and outlining future research directions. Notably, it explores leveraging variational inequalities to address and overcome a significant limitation in prior work that governs the strength of performative effects. This research aims to extend the analysis to scenarios with large performative effects and broaden the framework’s applicability, paving the way for more comprehensive solutions in performative prediction
Detektion, Quantifikation und Mitigation von Robustheitsanfälligkeiten in Tiefen Neuronalen Netzen
No full textMachine learning (ML) has made enormous progress in the last two decades. Specifically, Deep Neural Networks (DNNs) have led to several breakthroughs. The applications range from synthesizing high-resolution images that are indistinguishable from real photos to large-scale language models that achieve human-level performance on various tasks.
Yet, while humans can apply learned knowledge to new situations with only a few examples, neural networks often fail at this task. As a result, real-world distribution shifts in the environment, demographics, or data collection process pose severe safety risks to humans. For instance, autonomous cars may fail to adapt to unknown road conditions, and medical systems may provide incorrect diagnoses for minorities not included in the training data. Another security threat is the vulnerability of neural networks to small adversarially crafted perturbations. As such, even imperceptible changes in the data can lead to erroneous model behavior. In this cumulative dissertation, I demonstrate a literature gap regarding methods that simultaneously address real-world and adversarial distribution shifts. Therefore, I propose three objectives to increase the robustness of neural networks against both threats.
The first objective consists of the detection of potentially harmful model decisions caused by distribution shifts in the data. Here, we showed that the input-gradient geometry of neural networks can be used to detect both real-world and adversarial distribution shifts [P1]. Unlike prior work, we demonstrated the flexibility of our method by showing its effectiveness on both image and time series classification tasks.
The second objective considers the accurate quantification of network robustness against adversarial distribution shifts (attacks), which is essential to assess the worst-case risk in safety-critical applications. Toward this objective, we propose to improve two critical components of gradient-based adversarial attacks. In one contribution, we improved the convergence of gradient-descent-based optimization by including past gradient information in the optimization history [P2]. In another contribution, we introduced a novel optimization objective that leads to an increased attack success rate while simultaneously reducing the perturbation magnitude of adversarial attacks [P3].
Third, we present a novel approach to mitigate vulnerabilities against real-world and adversarial distribution shifts [P4]. To this end, we theoretically motivate how properties of local extrema in the loss landscape can be used to identify spurious predictions. Based on these findings, we propose the Decision Region Quantification (DRQ) algorithm that analyzes the robustness of local decision regions in the vicinity of a given data point to find the most robust prediction for a given sample
Learning Diverse Attacks on Large Language Models for Robust Red-Teaming and Safety Tuning
No full textL\u27Irlande de Georges Dor, Jacques Ferron et Louis Gauthier
No full textDans ce travail, l\u27auteur étudie trois romans d\u27auteurs québécois et aborde les liens qu\u27ils établissent entre l\u27Irlande et le Québec. L\u27étude s\u27intéressa aux rapprochements que l\u27on peut établir entre ces deux territoires et la manière dont ils sont représentés dans la littérature de Georges Dor, Jacques Ferron et Louis Gauthier et plus particulièrement dans les romans : Le fils de l\u27Irlandais, Le salut de l\u27Irlande et Voyage en Irlande avec un parapluie.
In this thesis, the author studies three books of Québécois authors and discusses the links that each one makes between Ireland and Quebec. This study looks at the relations that can be established between these two territories and the style in which they are represented in the works of Georges Dor, Jacques Ferron and Louis Gauthier, specifically in the books: Le fils de l\u27Irlandais, Le salut de l\u27Irlande and Voyage en Irlande avec un parapluie
Multi-player games in the era of machine learning
No full textParmi tous les jeux de société joués par les humains au cours de l’histoire, le jeu de go était considéré comme l’un des plus difficiles à maîtriser par un programme informatique [Van Den Herik et al., 2002]; Jusqu’à ce que ce ne soit plus le cas [Silveret al., 2016]. Cette percée révolutionnaire [Müller, 2002, Van Den Herik et al., 2002] fût le fruit d’une combinaison sophistiquée de Recherche arborescente Monte-Carlo et de techniques d’apprentissage automatique pour évaluer les positions du jeu, mettant en lumière le grand potentiel de l’apprentissage automatique pour résoudre des jeux. L’apprentissage antagoniste, un cas particulier de l’optimisation multiobjective, est un outil de plus en plus utile dans l’apprentissage automatique. Par exemple, les jeux à deux joueurs et à somme nulle sont importants dans le domain des réseaux génératifs antagonistes [Goodfellow et al., 2014] ainsi que pour maîtriser des jeux comme le Go ou le Poker en s’entraînant contre lui-même [Silver et al., 2017, Brown andSandholm, 2017]. Un résultat classique de la théorie des jeux indique que les jeux convexes-concaves ont toujours un équilibre [Neumann, 1928]. Étonnamment, les praticiens en apprentissage automatique entrainent avec succès une seule paire de réseaux de neurones dont l’objectif est un problème de minimax non-convexe et non-concave alors que pour une telle fonction de gain, l’existence d’un équilibre de Nash n’est pas garantie en général. Ce travail est une tentative d'établir une solide base théorique pour l’apprentissage dans les jeux. La première contribution explore le théorème minimax pour une classe particulière de jeux non-convexes et non-concaves qui englobe les réseaux génératifs antagonistes. Cette classe correspond à un ensemble de jeux à deux joueurs et a somme nulle joués avec des réseaux de neurones. Les deuxième et troisième contributions étudient l’optimisation des problèmes minimax, et plus généralement, les inégalités variationnelles dans le cadre de l’apprentissage automatique. Bien que la méthode standard de descente de gradient ne parvienne pas à converger vers l’équilibre de Nash de jeux convexes-concaves simples, il existe des moyens d’utiliser des gradients pour obtenir des méthodes qui convergent. Nous étudierons plusieurs techniques telles que l’extrapolation, la moyenne et la quantité de mouvement à paramètre négatif. La quatrième contribution fournit une étude empirique du comportement pratique des réseaux génératifs antagonistes. Dans les deuxième et troisième contributions, nous diagnostiquons que la méthode du gradient échoue lorsque le champ de vecteur du jeu est fortement rotatif. Cependant, une telle situation peut décrire un pire des cas qui ne se produit pas dans la pratique. Nous fournissons de nouveaux outils de visualisation afin d’évaluer si nous pouvons détecter des rotations dans comportement pratique des réseaux génératifs antagonistes.Among all the historical board games played by humans, the game of go was considered one of the most difficult to master by a computer program [Van Den Heriket al., 2002]; Until it was not [Silver et al., 2016]. This odds-breaking break-through [Müller, 2002, Van Den Herik et al., 2002] came from a sophisticated combination of Monte Carlo tree search and machine learning techniques to evaluate positions, shedding light upon the high potential of machine learning to solve games. Adversarial training, a special case of multiobjective optimization, is an increasingly useful tool in machine learning. For example, two-player zero-sum games are important for generative modeling (GANs) [Goodfellow et al., 2014] and mastering games like Go or Poker via self-play [Silver et al., 2017, Brown and Sandholm,2017]. A classic result in Game Theory states that convex-concave games always have an equilibrium [Neumann, 1928]. Surprisingly, machine learning practitioners successfully train a single pair of neural networks whose objective is a nonconvex-nonconcave minimax problem while for such a payoff function, the existence of a Nash equilibrium is not guaranteed in general. This work is an attempt to put learning in games on a firm theoretical foundation. The first contribution explores minimax theorems for a particular class of nonconvex-nonconcave games that encompasses generative adversarial networks. The proposed result is an approximate minimax theorem for two-player zero-sum games played with neural networks, including WGAN, StarCrat II, and Blotto game. Our findings rely on the fact that despite being nonconcave-nonconvex with respect to the neural networks parameters, the payoff of these games are concave-convex with respect to the actual functions (or distributions) parametrized by these neural networks. The second and third contributions study the optimization of minimax problems, and more generally, variational inequalities in the context of machine learning. While the standard gradient descent-ascent method fails to converge to the Nash equilibrium of simple convex-concave games, there exist ways to use gradients to obtain methods that converge. We investigate several techniques such as extrapolation, averaging and negative momentum. We explore these techniques experimentally by proposing a state-of-the-art (at the time of publication) optimizer for GANs called ExtraAdam. We also prove new convergence results for Extrapolation from the past, originally proposed by Popov [1980], as well as for gradient method with negative momentum. The fourth contribution provides an empirical study of the practical landscape of GANs. In the second and third contributions, we diagnose that the gradient method breaks when the game’s vector field is highly rotational. However, such a situation may describe a worst-case that does not occur in practice. We provide new visualization tools in order to exhibit rotations in practical GAN landscapes. In this contribution, we show empirically that the training of GANs exhibits significant rotations around Local Stable Stationary Points (LSSP), and we provide empirical evidence that GAN training converges to a stable stationary point, which is a saddle point for the generator loss, not a minimum, while still achieving excellent performance
Guerre du blé, guerre du droit
No full textWhat the historiography of the French Revolution as forgotten is the reflexion on the rights of men, the first of which is the one which guarantees life. Against an economist and determinist vision, the author re-estimates the place of the peasantry in the process of liberation of 1789.Gauthier Florence, EspaceTemps. Guerre du blé, guerre du droit. In: Espaces Temps, 38-39, 1988. Concevoir la révolution. 89, 68, confrontations. pp. 113-114
The contextual database of the generations and gender program in Bulgaria: conceptual framework and an overview of the Bulgarian context concerning the central database topics
No full textThis paper outlines the concept and content of the Contextual Database of the international Generations and Gender Program and gives an overview of the context of demographic behavior in Bulgaria. The Contextual Database provides an instrument that together with the Generations and Gender Survey allows studying how differences in context shape demographic processes. The database offers the opportunity to analyze in a comparative way the interaction between the micro and macro dimension. Bulgaria is among the first countries fielding the Generations and Gender Survey and that is engaged in contextual data collection within this comparative framework. While both micro- and contextual data for Bulgaria will become available in the course of the year 2005, we present in this paper a text contribution that provides an overview of the Bulgarian context and introduces the list of variables that make up the database.Bulgaria, data collection
RabbitStamp Test Sequence
No full text# RabbitStamp sequence by LISA ULB
The test sequence "RabbitStamp" is provided by Sarah Fachada, Yupeng Xie, Daniele Bonatto, Gauthier Lafruit, Mehrdad Teratani, members of the LISA department, EPB (Ecole Polytechnique de Bruxelles), ULB (Universite Libre de Bruxelles), Belgium.
# License:
CC BY-NC-SA
ONLY Available for Academic Usage
# Terms of Use:
Anykind of publication or report using this sequence should refer to the following references.
[1] Sarah Fachada, Yupeng Xie, Daniele Bonatto, Gauthier Lafruit, Mehrdad Teratani, "RabbitStamp Test Sequence", 2021.
@misc{fachada_RabbitStamp_2021,
title = {{RabbitStamp} {Test} {Sequence}},
author = {Fachada, Sarah and Xie; Yupeng and Bonatto, Daniele and Lafruit, Gauthier and Teratani, Mehrdad },
month = jul,
year = {2021},
doi = {10.5281/zenodo.5053771}
}
[2] Sarah Fachada, Yupeng Xie, Daniele Bonatto, Gauthier Lafruit, Mehrdad Teratani, "[DLF] Plenoptic 2.0 Multiview Lenslet Dataset and Preliminary Experiments [m56429]", 2021.
@article{fachada_RabbitStamp_2021,
title = {[DLF] {Plenoptic} 2.0 {Multiview} {Lenslet} {Dataset} and {Preliminary} {Experiments} [m56429]},
author = {Fachada, Sarah and Xie; Yupeng and Bonatto, Daniele and Lafruit, Gauthier and Teratani, Mehrdad },
journal = {ISO/IEC JTC1/SC29/WG11},
month = apr,
year = {2021}
}
[3] Sarah Fachada, Yupeng Xie, Daniele Bonatto, Gauthier Lafruit, Mehrdad Teratani, "[LVC] Update for RabbitStamp: Plenoptic 2.0 Multiview Lenslet Dataset [m57100]", 2021.
@article{fachada_RabbitStamp_2021,
title = {[LVC] {Update} for {RabbitStamp}: {Plenoptic} 2.0 {Multiview} {Dataset} [m56429]},
author = {Fachada, Sarah and Xie; Yupeng and Bonatto, Daniele and Lafruit, Gauthier and Teratani, Mehrdad },
journal = {ISO/IEC JTC1/SC29/WG11},
month = jul,
year = {2021}
}
[4] Sarah Fachada, Yupeng Xie, Daniele Bonatto, Gauthier Lafruit, Mehrdad Teratani, "[LVC] Exploration Experiments using RabbitStamp Multiview Lenslet Images [m57101]", 2021.
@article{fachada_RabbitStamp_2021,
title = {[LVC] {Exploration} {Experiments} {Using} {RabbitStamp} {Multiview} {Lenslet} {Images} [m56429]},
author = {Fachada, Sarah and Xie; Yupeng and Bonatto, Daniele and Lafruit, Gauthier and Teratani, Mehrdad},
journal = {ISO/IEC JTC1/SC29/WG11},
month = jul,
year = {2021}
}
# Production:
Laboratory of Image Synthesis and Analysis, LISA department, Ecole Polytechnique de Bruxelles, Universite Libre de Bruxelles, Belgium.
# Content:
This dataset contains a test scene acquired with a raytrix camera [1] array of 7x3 views. For details of the dataset, please refer to the references mentioned above.
The dataset contains:
- a `depth_7x3_center` depth maps computed with DERS reference software [2] in yuv40016ble format and json configuration files to do so,
- a `multiview_7x3_5x5_images` Calibrated subimages computed with RLC [3] in yuv42010ble format, the cameras.json with the camera parameters and view_synthesis.json with the view synthesis experiment.
- a `multiview_7x3_lenslets` folder containing the lenslet views in yuv42010ble format, the Raytrix xml calibration file and RLC cfg file for conversion to multiview.
# References and links:
[1] Raytrix, https://raytrix.de/
[2] S. Rogge and D. Bonatto and J. Sancho and R. Salvador and E. Juarez and A. Munteanu and G. Lafruit, "MPEG-I Depth Estimation Reference Software", in 2019 International Conference on 3D Immersion (IC3D), 2019.
[3] M. Teratani and T. Fujii, "[MPEG-I Visual] Conversion of Lenslet Data Capture by Single Focussed Plenoptic Camera to Multiview Video using RLC0.3 [N18567]", ISO/IEC JTC1/SC29/WG11, 201
Le constructivisme
No full textGilles Gauthier réplique aux différentes réactions suscitées par sa critique du constructivisme en journalisme. Il commence par caractériser cette notion par deux thèses fondamentales : une thèse épistémologique – l’anti-objectivisme cognitif – posant que la connaissance ne relève pas d’une adéquation au réel ; une thèse méta-ontologique – le scepticisme ontologique – qui appelle à une suspension du jugement sur l’existence de la réalité. Il soutient ensuite que le constructivisme, ainsi caractérisé, est incompatible avec le journalisme en établissant : le rapport nécessaire du journalisme à la réalité ; la dépendance de la construction journalistique à l’égard d’un donné ; la possibilité de produire des énoncés objectifs en journalisme ; l’assujettissement du journalisme à une norme éthique de vérité.Gilles Gauthier responds to the reactions generated by his analysis of constructivism in journalism. He will begin by characterizing constructivism by two fundamental theses: the epistemological perspective or cognitive anti-objectivism, which suggests that knowledge is not a matter of suitability to reality, and the meta-ontological perspective or ontological scepticism which calls for a suspended judgment with regard to the existence of reality. The author will then discuss this constructivism’s incompatibility with journalism by establishing that: journalism is necessarily linked to reality; journalistic construction depends on a given reality; objective discourse in journalism is possible; journalism follows an ethical standard of reality
- …
