Gotowa bibliografia na temat „Exemples contradictoires”
Utwórz poprawne odniesienie w stylach APA, MLA, Chicago, Harvard i wielu innych
Spis treści
Zobacz listy aktualnych artykułów, książek, rozpraw, streszczeń i innych źródeł naukowych na temat „Exemples contradictoires”.
Przycisk „Dodaj do bibliografii” jest dostępny obok każdej pracy w bibliografii. Użyj go – a my automatycznie utworzymy odniesienie bibliograficzne do wybranej pracy w stylu cytowania, którego potrzebujesz: APA, MLA, Harvard, Chicago, Vancouver itp.
Możesz również pobrać pełny tekst publikacji naukowej w formacie „.pdf” i przeczytać adnotację do pracy online, jeśli odpowiednie parametry są dostępne w metadanych.
Artykuły w czasopismach na temat "Exemples contradictoires"
Rhéault, Sylvie. "Domicile ou hébergement? Quand les croyances prennent le dessus". Service social 43, nr 1 (12.04.2005): 33–46. http://dx.doi.org/10.7202/706641ar.
Pełny tekst źródłaGraham, Kenneth J. E. ""Clear as heav'n:" Herbert's Poetry and Rhetorical "Divinitie"". Renaissance and Reformation 41, nr 2-3 (1.01.2005): 183–201. http://dx.doi.org/10.33137/rr.v41i2-3.9528.
Pełny tekst źródłaDuchastel, Jules, i Danielle Laberge. "Transformation des modes étatiques de contrôle social". Politique, nr 20 (10.12.2008): 65–92. http://dx.doi.org/10.7202/040699ar.
Pełny tekst źródłaBinhas, Edmond, i Corinne Binhas. "Gérer son cabinet d’orthodontie entre rigueur et flexibilité". L'Orthodontie Française 88, nr 1 (23.02.2017): 81–86. http://dx.doi.org/10.1051/orthodfr/2017003.
Pełny tekst źródłaClavier, Tatiana. "La Querelle des femmes au coeur de quatre « Institutions » de l’élite imprimées et largement diffusées en français dans la seconde moitié du XVIe siècle". Renaissance and Reformation 46, nr 3 (15.02.2024): 163–89. http://dx.doi.org/10.33137/rr.v46i3.42640.
Pełny tekst źródłaBerry, Sara. "Hegemony on a shoestring: indirect rule and access to agricultural land". Africa 62, nr 3 (lipiec 1992): 327–55. http://dx.doi.org/10.2307/1159747.
Pełny tekst źródłaMottier, Véronique. "État et contrôle de la sexualité reproductive : l’exemple des politiques eugénistes dans les démocraties libérales (Suisse, Suède et Royaume-Uni)1". Articles 31, nr 2 (22.02.2013): 31–50. http://dx.doi.org/10.7202/1014350ar.
Pełny tekst źródłaKassühlke, Rudolph. "Gedanken zur Übersetzung poetischer Bibeltexte". Meta 32, nr 1 (30.09.2002): 76–84. http://dx.doi.org/10.7202/003284ar.
Pełny tekst źródłaINGRAND, S., i B. DEDIEU. "Numéro spécial : Quelles innovations pour quels systèmes d'élevage ?" INRAE Productions Animales 27, nr 2 (1.06.2014): 75–76. http://dx.doi.org/10.20870/productions-animales.2014.27.2.3055.
Pełny tekst źródłaBeauregard, Micheline. "Féminisme et sémiotique greimassienne : deux problématiques aussi irréconciliables que réalité et fiction?" Notes de recherche 2, nr 2 (12.04.2005): 147–56. http://dx.doi.org/10.7202/057564ar.
Pełny tekst źródłaRozprawy doktorskie na temat "Exemples contradictoires"
Guiga, Linda. "Software protections for artificial neural networks". Electronic Thesis or Diss., Institut polytechnique de Paris, 2022. http://www.theses.fr/2022IPPAT024.
Pełny tekst źródłaIn a context where Neural Networks (NNs) are very present in our daily lives, be it through smartphones, face and biometrics recognition or even in the medical field, their security is of the utmost importance. If such models leak information, not only could it imperil the privacy of sensitive data, but it could also infringe on intellectual property.Selecting the correct architecture and training the corresponding parameters is time-consuming -- it can take months -- and requires large computational resources. This is why an NN constitutes intellectual property. Moreover, once a malicious user knows the architecture and/or the parameters, multiple attacks can be carried out, such as adversarial ones. Adversarial attackers craft a malicious datapoint by adding a small noise to the original input, such that the noise is undetectable to the human eye but fools the model. Such attacks could be the basis of impersonations. Membership attacks, which aim at leaking information about the training dataset, are also facilitated by the knowledge of a model. More generally, when a malicious user has access to a model, she also has access to the manifold of the model's outputs, making it easier for her to fool the model.Protecting NNs is therefore paramount. However, since 2016, they have been the target of increasingly powerful reverse-engineering attacks. Mathematical reverse-engineering attacks solve equations or study a model's internal structure to reveal its parameters. On the other hand, side-channel attacks exploit leaks in a model's implementation -- such as in the cache or power consumption -- to uncover the parameters and architecture. In this thesis, we seek to protect NN models by changing their internal structure and their software implementation.To this aim, we propose four novel countermeasures. In the first three, we consider a gray-box context where the attacker has partial access to the model, and we leverage parasitic models to counter three types of attacks.We first tackle mathematical attacks that recover a model's parameters based on its internal structure. We propose to add one -- or multiple -- parasitic Convolutional Neural Networks (CNNs) at various locations in the base model and measure the incurred change in the structure by observing the modification in generated adversarial samples.However, the previous method does not thwart side-channel attacks that extract the parameters through the analysis of power or electromagnetic consumption. To mitigate such attacks, we propose to add dynamism to the previous protocol. Instead of considering one -- or several -- fixed parasite(s), we incorporate different parasites at each run, at the entrance of the base model. This enables us to hide a model's input, necessary for precise weight extraction. We show the impact of this dynamic incorporation through two simulated attacks.Along the way, we observe that parasitic models affect adversarial examples. Our third contribution is derived from this, as we suggest a novel method to mitigate adversarial attacks. To this effect, we dynamically incorporate another type of parasite: autoencoders. We demonstrate the efficiency of this countermeasure against common adversarial attacks.In a second part, we focus on a black-box context where the attacker knows neither the architecture nor the parameters. Architecture extraction attacks rely on the sequential execution of NNs. The fourth and last contribution we present in this thesis consists in reordering neuron computations. We propose to compute neuron values by blocks in a depth-first fashion, and add randomness to this execution. We prove that this new way of carrying out CNN computations prevents a potential attacker from recovering a small enough set of possible architectures for the initial model
Hamza, Anis Amazigh. "Improving cooperative non-orthogonal multiple access (CNOMA) and enhancing the physical layer security (PLS) for beyond 5G (B5G) and future eHealth wireless networks". Electronic Thesis or Diss., Valenciennes, Université Polytechnique Hauts-de-France, 2023. http://www.theses.fr/2023UPHF0006.
Pełny tekst źródłaThe fifth generation of cellular networks (5G) was a real revolution in radio access technologies and mobile networks, presenting itself as the breakthrough generation that allowed the coexistence of extremely diversified applications and usage scenarios, unified under the same standard. Nevertheless, 5G is just the beginning: new scenarios and challenges are emerging. Therefore, the research community is pushing the research ahead and preparing the ground for beyond 5G (B5G) cellular systems. In this regard, several enabling technologies are investigated. In addition to the cognitive radio (CR), mmWave, massive MIMO, or even the use of full-duplex (FD), non-orthogonal multiple access (NOMA) emerged as a promising technology that allows multiple users to share the same resource block and hence, optimizes resource allocation, reduces the end-to-end latency, and improves both spectrum and energy efficiencies. Those advantages make NOMA a serious candidate as a multiple access scheme for future B5G networks, especially for the demanding eHealth applications. Furthermore, NOMA can be flexibly combined with any wireless technology such as cooperative communication, FD, mmWave, and multicarrier modulation (MCM).Motivated by this treatise, this thesis provides a comprehensive and intensive examination of this emerging technology, particularly, cooperative NOMA (CNOMA) which is considered a promising enabling technology for future B5G eHealth networks, from the basic principles to its combination with the full-duplex technology, MCM transmission, to deep learning as well as enhancing the physical layer security (PLS).First, this thesis investigates the error rate performance of FD-CNOMA systems over wireless fading channels. New closed-form expressions of the exact bit error rates (BER) are derived. Moreover, high-SNR analyses are conducted, which reveals that FD-CNOMA has an error floor due to the successive interference cancellation (SIC) imperfections and residual self-interference (RSI). Based on the derived expressions, a novel selective relaying scheme is proposed to opportunistically improve the system performance using the minimal channel state information (CSI) overhead.Second, the MCM-based CNOMA is examined under doubly selective channels encountered in vehicular and railway wireless communications. In the eHealth context, this can be projected to ambulance emergency healthcare use cases. More importantly, this thesis presents a performance improvement method for cell-edge users of MCM-NOMA systems with imperfect SIC and imperfect CSI under doubly selective wireless channels. Two efficient iterative interference cancellation schemes are proposed to enable user relaying for MCM-based CNOMA. The proposed schemes are robust for high mobility scenarios with a relatively low computational complexity.Third and last, advances in machine learning based on deep neural networks (DNNs) attracted great attention in the wireless communication community (WCS). It is regarded as a key component of B5G networks. Deep learning has found a broad range of applications in wireless systems, e.g., spectrum sensing, waveform design, SIC, and channel estimation. However, DNNs are known to be highly susceptible to adversarial attacks. Many robust over-the-air adversarial attacks against DNN-based WCS have been proposed in the literature. This is becoming a major challenge facing the physical layer security (PLS) of DNN-based WCS. To overcome this vulnerability, this thesis proposes a novel robust defense approach. The objective of our defense is to protect the victim without significantly degrading the accuracy of its baseline model in the absence of the attack. The obtained results are very promising and confirm that the proposed defense technique can enhance significantly the PLS of future DNN-based WCS