4,647 research outputs found
TRIBUNAL ARBITRAL DE PAULA ANDREA SANCHEZ CEBALLOS VS. CENTRAL DE INVERSIONES S.A -C. I. S. A.
No full text13079
How to Approximate any Objective Function via Quadratic Unconstrained Binary Optimization
No full textQuadratic unconstrained binary optimization (QUBO) has become the standard format for optimization using quantum computers, i.e., for both the quantum approximate optimization algorithm (QAOA) and quantum annealing (QA). We present a toolkit of methods to transform almost arbitrary problems to QUBO by (i) approximating them as a polynomial and then (ii) translating any polynomial to QUBO. We showcase the usage of our approaches on two example problems (ratio cut and logistic regression).Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Quantum Circuit Architectures and Technolog
Structured Test Development Approach for Computation-in-Memory Architectures
No full textTesting of Computation-in-Memory (CIM) designs based on emerging non-volatile memory technologies, such as resistive RAM (RRAM), is fundamentally different from testing traditional memories. Such designs allow not only for data storage (i.e., memory configuration) but also for the execution of logical and arithmetic operations (i.e., computing configuration). Therefore, not only significant design changes are needed in the memory array and/or in the peripheral circuits, but also new fault models and test approaches are needed. Moreover, RRAM-based CIM makes use of non-linear non-volatile devices making the defect modeling with traditional linear resistor inappropriate for such device defects. Hence, even the way of doing defect modeling has to change. This paper discusses a structured test development approach for RRAM-based CIM and highlights the test challenges and how testing CIM dies is different from the traditional way of testing logic and memory. Methods for defect modeling, fault modeling, and test development will be discussed. The paper demonstrates that unique faults can occur in the CIM die while in the computation configuration and that these faults cannot be detected by just testing the CIM die in the memory configuration. Moreover, it shows that testing the CIM die in the computation configuration reduces the overall test time while improving the outgoing product quality. Finally, the paper presents an outlook on the future of structured CIM test development.Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Quantum & Computer EngineeringComputer Engineerin
Improved DQN-Based Computation Offloading Algorithm in MEC Environment
No full textMassive terminal users have brought explosive need of data residing at edge of overall network. Multiple Mobile Edge Computing (MEC) servers are built in/near base station to meet this need. However, optimal distribution of these servers to multiple users in real time is still a problem. Reinforcement Learning (RL) as a framework to solve interaction problem is a promising solution. In order to apply RL based algorithm into a multi-agent environment, we propose an iterative scheme: select individual users with priorities to interact with the environment iteratively one at a time Furthermore, we tried to optimize the overall system performance based on this scheme. Hence, we construct three objective system performance indicators: average processing cost, delay and energy consumption, improve the existing Deep Q-learning Network (DQN) by using the cost as reward function, changing the fixed exploitation rate into dynamic one that associated with reward and episode time. In order to explore the performance potential of the proposed algorithm, we have simulated the proposed algorithm, DQN algorithm and greedy algorithm under different users and data sizes. The results show that the proposed algorithm had reduced at least 12% of system average processing cost comparing to the greedy algorithm. It also outperform the greedy algorithm and DQN algorithm in delay and energy consumption significantly.Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Computer Engineerin
Removing dependencies from large software projects: Are you really sure?
No full textWhen developing and maintaining large software systems, a great deal of effort goes into dependency management. During the whole lifecycle of a software project, the set of dependencies keeps changing to accommodate the addition of new features or changes in the running environment. Package management tools are quite popular to automate this process, making it fairly easy to automate the addition of new dependencies and respective versions. However, over the years, a software project might evolve in a way that no longer needs a particular technology or dependency. But the choice of removing that dependency is far from trivial: one cannot be entirely sure that the dependency is not used in any part of the project. Hence, developers have a hard time confidently removing dependencies and trusting that it will not break the system in production. In this paper, we propose a decision framework to improve the detection of unused dependencies. Our approach builds on top of the existing dependency analysis tool DepClean. We start by improving the support of Java dynamic features in DepClean. We do so by augmenting the analysis with the state-of-the-art call graph generation tool OPAL. Then, we analyze the potentially unused dependencies detected by classifying their logical relationship with the other components to decide on follow-up steps, which we provide in the form of a decision diagram. Results show that developers can focus their efforts on maintaining bloated dependencies by following the recommendations of our decision framework. When applying our approach to a large industrial software project, we can reduce one-third of false positives when compared to the state-of-the-art. We also validate our approach by analyzing dependencies that were removed in the history of open-source projects. Results show consistency between our approach and the decisions taken by open-source developers.Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Software EngineeringSoftware Technolog
AGIC: Approximate Gradient Inversion Attack on Federated Learning
No full textFederated learning is a private-by-design distributed learning paradigm where clients train local models on their own data before a central server aggregates their local updates to compute a global model. Depending on the aggregation method used, the local updates are either the gradients or the weights of local learning models, e.g., FedAvg aggregates model weights. Unfortunately, recent reconstruction attacks apply a gradient inversion optimization on the gradient update of a single mini- batch to reconstruct the private data used by clients during training. As the state-of-the-art reconstruction attacks solely focus on single update, realistic adversarial scenarios are over- looked, such as observation across multiple updates and updates trained from multiple mini-batches. A few studies consider a more challenging adversarial scenario where only model updates based on multiple mini-batches are observable, and resort to computationally expensive simulation to untangle the underlying samples for each local step. In this paper, we propose AGIC, a novel Approximate Gradient Inversion Attack that efficiently and effectively reconstructs images from both model or gradient updates, and across multiple epochs. In a nutshell, AGIC (i) approximates gradient updates of used training samples from model updates to avoid costly simulation procedures, (ii) leverages gradient/model updates collected from multiple epochs, and (iii) assigns increasing weights to layers with respect to the neural network structure for reconstruction quality. We extensively evaluate AGIC on three datasets, namely CIFAR-10, CIFAR- 100 and ImageNet. Our results show that AGIC increases the peak signal-to-noise ratio (PSNR) by up to 50% compared to two representative state-of-the-art gradient inversion attacks. Furthermore, AGIC is faster than the state-of-the-art simulation- based attack, e.g., it is 5x faster when attacking FedAvg with 8 local steps in between model updates.Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Data-Intensive System
Secure Embedding of Rooted Spanning Trees for Scalable Routing in Topology-Restricted Networks
No full textGreedy embeddings on rooted spanning trees are the most promising solution to provide sufficiently scalable routing in dynamic networks with restricted topologies, for instance friend-to-friend overlays such as the Dark Freenet and payment channel networks such as Lightning. Yet, they are not deployed in practice, as electing a root and configuring addresses remains an unsolved problem in adverse environments. Indeed, faulty or malicious nodes might provide incorrect coordinates, prevent the network from stabilizing by simulating dynamics, or not start the assignment of coordinates in their subtree at all. All of the above attacks may result in an inability to route. To mitigate the above attacks, we design a novel embedding algorithm with an adapted distance metric that only relies on interconnections between benign subtrees for successful delivery. In other words, even if roots of (sub-)trees are malicious or faulty, the remaining nodes still receive coordinates and can communicate with nodes in their tree branch as well as other branches reachable via the neighborhood of their benign ancestors. Extensive simulations demonstrate that we thus facilitate efficient routing even when seemingly decisive parts of the network are under adversarial control.Data-Intensive System
Pattern of ivermectin (sheep) and doramectin (cattle) residues in muscular tissue from various anatomical locations
No full textThis trial reports comparative drug residual concentrations in muscular tissue obtained from various anatomical locations after subcutaneous administration of ivermectin (IVM) to sheep and topical treatment with doramectin (DRM) to calves at recommended therapeutic dose rates. Seven muscle samples from different anatomical locations (rhomboideus, supraspinatus, semitendinosus, gluteus medius, longissimus dorsi thoracis, intercostales and diaphragma) were collected at several post-treatment sampling times. Samples were frozen at-20 degrees C until analyzed by HPLC. The highest IVM residual concentrations in muscular tissue from the different locations were found at 15 days post-treatment in sheep. Although the highest IVM mean concentrations were measured at 15 (16.8 +/- 5.17 ng g(-1)) and 20 (10.5 +/- 4.06 ng g(-1)) days post-administration in the intercostales muscles, at 30 days post-administration, the IVM concentration in this location was similar to that measured in the rhomboideus and diaphragma muscles. DRM residual concentrations were quantified in muscular tissue from all anatomical locations after topical administration to calves. Maximum residue level was observed at 10 days post-treatment in all anatomical sites. The diaphragma muscle showed the highest DRM residue levels at 2 (22.0 +/- 4.35 ng g(-1)), 5 (45.2 +/- 3.78 ng g(-1)) and 10 (57.9 +/- 9.57 ng g(-1)) days post-treatment in calves. These results demonstrated that the pattern of residue depletion from muscular tissue may differ according to its anatomical locations and/or physiological role. This should be considered in implementing residue control strategies in meat safety assurance for human consumption.Fil: Moreno Torrejon, Laura. Universidad Nacional del Centro de la Provincia de Buenos Aires. Facultad de Ciencias Veterinarias. Departamento de Fisiopatología. Laboratorio de Farmacología; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; ArgentinaFil: Alvarez, Luis Ignacio. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Universidad Nacional del Centro de la Provincia de Buenos Aires. Facultad de Ciencias Veterinarias. Departamento de Fisiopatología. Laboratorio de Farmacología; ArgentinaFil: Ceballos, Laura. Universidad Nacional del Centro de la Provincia de Buenos Aires. Facultad de Ciencias Veterinarias. Departamento de Fisiopatología. Laboratorio de Farmacología; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; ArgentinaFil: Sanchez Bruni, Sergio Fabian. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Universidad Nacional del Centro de la Provincia de Buenos Aires. Facultad de Ciencias Veterinarias. Departamento de Fisiopatología. Laboratorio de Farmacología; ArgentinaFil: Lanusse, Carlos Edmundo. Universidad Nacional del Centro de la Provincia de Buenos Aires. Facultad de Ciencias Veterinarias. Departamento de Fisiopatología. Laboratorio de Farmacología; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentin
Proprotein convertase subtilisin/kexin type 9 inhibitors treatment in dyslipidemic patients: a real world prescription
No full textAim Dyslipidemia is recognized as one of the major risk factors for cardiovascular diseases. This retrospective observational study was aimed to assess the effect of proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors in dyslipidemic patients with a lipid profile not well controlled by maximally tolerated statin therapy or intolerant to these lipid-lowering drugs. We enrolled 151 patients, of whom, 119 were taking evolocumab and 32 alirocumab. Results Total cholesterol significantly decreased progressively until the fourth year; after 4 years there was a significant reduction (S125.5 mg/dl, S51.5%, P < 0.0001 vs baseline, and P < 0.05 vs 1 year and P < 0.05 vs 2 years) and S2.8 mg/dl (S2.3%) compared with the third year. Low-density lipoprotein-cholesterol (LDL-C) also decreased significantly until the fourth year. After 3 years, there was a significant reduction (S117.8 mg/dl, S71.5%, P < 0.0001 vs baseline, and P < 0.05 vs 1 year) and S13.9 mg/dl (S22.8%) compared with the second year; after 4 years there was a significant reduction (S121.4 mg/dl, S73.7%, P < 0.0001 vs baseline, and P < 0.05 vs 1 year and P < 0.05 vs 2 years) and S3.6 mg/dl (S7.7%) compared with the third year. High-density lipoprotein-cholesterol increased significantly only during the fourth year of detection. After 3 years, there was a nonsignificant increase (4.9 mg/dl, 10.0%, P U 0.061 vs baseline) and 1.6 mg/dl (3.1%) compared with the second year; after 4 years, there was a significant increase (5.2 mg/dl, 10.6%, P < 0.05 vs baseline) and 0.3 mg/dl (0.6%) compared with the third year. The value of Tg was significantly reduced progressively until the second year and then stabilized in the third and fourth years. After 3 years, the value of Tg stabilized (S48.6 mg/dl, S32.4%, P < 0.01 vs baseline, and P < 0.05 vs 1 year) and S4.8 mg/dl (S4.5%) compared with the second year; after 4 years (S46.4 mg/dl, S31.0%, P < 0.01 vs baseline, and P < 0.05 vs 1 year) there was a slight and nonsignificant increase of 2.2 mg/dl (2.2%) compared with the third year. Regarding adverse events, both drugs were well tolerated. Conclusions We showed that PCSK9 inhibitors are well tolerated and provide long-term significant LDL-C lowering in individuals with hyperlipidemia
Thermal stability of structures such as TiN/ZrO2/InGaAs
No full textLes semiconducteurs composés III-V, et en particulier l’InGaAs, sont considéréscomme une alternative attractive pour remplacer le Silicium (Si) habituellement utilisépour former le canal dans les dispositifs Métal-Oxide-Semiconducteur (MOS). Sa hautemobilité électronique et sa bande interdite modulable, des paramètres clés pourl’ingénierie de dispositifs à haute performance, ont fait de l’InGaAs un candidatprometteur. Cependant, la stabilité thermique et la chimie des interfaces desdiélectriques high-k sur InGaAs est beaucoup plus complexe que sur Si. Tandis que laplupart des études se concentrent sur diverses méthodes de passivation, telles que lacroissance de couches passivantes d’interface (Si, Ge, et Si/Ge) et/ou le traitementchimique afin d’améliorer la qualité de l’interface high-k/InGaAs, les phénomènes telsque la diffusion d’espèces atomiques provenant du substrat dus aux traitementsthermiques n’ont pas été étudiés attentivement. Les traitements thermiques liés auxprocédés d’intégration de la source (S) et du drain (D) induisent des changementsstructurels qui dégradent les performances électriques du dispositif MOS. Unecaractérisation adaptée des altérations structurelles associées à la diffusion d’élémentsdepuis la surface du substrat est importante afin de comprendre les mécanismes defaille. Dans ce travail, une analyse de la structure ainsi que de la stabilité thermiquedes couches TiN/ZrO2/InGaAs par spectroscopie de photoélectrons résolue en angle(ARXPS) est présentée. Grâce à cette méthode d’analyse non destructive, il a étépossible d’observer des effets subtils tels que la diffusion d’espèces atomiques àtravers la couche diélectrique due au recuit thermique. A partir de la connaissance dela structure des couches, les profils d’implantation d’In et de Ga ont pu être estiméspar la méthode des scenarios. L’analyse de l’échantillon avant recuit thermique apermis de localiser les espèces In-O et Ga-O à l’interface oxide-semiconducteur. Aprèsrecuit, les résultats démontrent de façon quantitative que le recuit thermique cause ladiffusion de In et Ga vers les couches supérieures. En considérant différents scénarios,il a pu être démontré que la diffusion d’In et de Ga induite par le recuit atteint lacouche de TiO2. Dans le cas où l’échantillon est recuit à 500 °C, seule la diffusion d’Inest clairement observée, tandis que dans le cas où l’échantillon est recuit à 700 °C, onobserve la diffusion d’In et de Ga jusqu’à la couche de TiO2. L’analyse quantitative~ viii ~montre une diffusion plus faible de Gallium (~ 0.12 ML) que d’Indium (~ 0.26 ML) à 700°C /10 s. L’analyse quantitative en fonction de la température de recuit a permisd’estimer la valeur de l’énergie d’activation pour la diffusion d’Indium à travers leZircone. La valeur obtenue est très proche des valeurs de diffusion de l’Indium àtravers l’alumine et l’hafnia précédemment rapportées. Des techniquescomplémentaires telles que la microscopie électronique en transmission à hauterésolution (HR-TEM), la spectroscopie X à dispersion d’énergie (EDX) et laspectrométrie de masse à temps de vol (TOF-SIMS) ont été utilisés pour corréler lesrésultats obtenus par ARXPS. En particulier, la TOF-SIMS a révélé le phénomène dediffusion des espèces atomiques vers la surface.III-V compound semiconductors, in particular InGaAs, are considered attractivealternative channel materials to replace Si in complementary metal-oxidesemiconductor(MOS) devices. Its high mobility and tunable band gap, requirementsfor high performance device design, have placed InGaAs as a promising candidate.However, the interfacial thermal stability and chemistry of high-k dielectrics on InGaAsis far more complex than those on Si. While most studies are focused on variouspassivation methods, such as the growth of interfacial passivation layers (Si, Ge, andSi/Ge) and/or chemical treatments to improve the quality of high-k/InGaAs interface,phenomena such as the out-diffusion of atomic species from the substrate as aconsequence of the thermal treatments have not been carefully studied. The thermaltreatments, which are related with integration processes of source and drain (S/D),lead to structural changes that degrade the electrical performance of the MOS device.A proper characterization of the structural alterations associated with the out-diffusionof elements from the substrate is important for understanding failure mechanisms. Inthis work it is presented an analysis of the structure and thermal stability ofTiN/ZrO2/InGaAs stacks by angle-resolved x-ray photoelectron spectroscopy (ARXPS).Through a non-destructive analysis method, it was possible to observe subtle effectssuch as the diffusion of substrate atomic species through the dielectric layer as aconsequence of thermal annealing. The knowledge of the film structure allowed forassessing the In and Ga depth profiles by means of the scenarios-method. For the asdeposited sample, In-O and Ga-O are located at the oxide-semiconductor interface. Byassuming different scenarios for their distribution, it was quantitatively shown thatannealing causes the diffusion of In and Ga up to the TiO2 layer. For the sampleannealed at 500 °C, only the diffusion of indium was clearly observed, while for thesample annealed at 700 °C the diffusion of both In and Ga to the TiO2 layer wasevident. The quantitative analysis showed smaller diffusion of gallium (~ 0.12 ML) thanof indium (~ 0.26 ML) at 700 °C/10 s. Since the quantification was done at differenttemperatures, it was possible to obtain an approximate value of the activation energyfor the diffusion of indium through zirconia. The value resulted to be very similar topreviously reported values for indium diffusion through alumina and through hafnia.~ vi ~Complementary techniques as high resolution transmission electron microscopy (HRTEM),energy dispersive x-ray spectroscopy (EDX) and time of flight secondary ion massspectrometry (TOF-SIMS) were used to complement the results obtained with ARXPS.Specially, TOF-SIMS highlighted the phenomenon of diffusion of the substrate atomicspecies to the surface.Compuestos semiconductores III-V, en particular InxGa1-xAs, son consideradosmateriales atractivos para reemplazar el silicio en estructuras metal-oxidosemiconductor(MOS). Su alta movilidad y flexible ancho de banda, requisitos para eldiseño de dispositivos de alto rendimiento, han colocado al InxGa1-xAs como uncandidato prometedor. Sin embargo, la estabilidad térmica en la interfazdieléctrico/InxGa1-xAs es mucho más compleja que aquella formada en la estructuraSiO2/Si. Mientras que la mayoría de los estudios se centran en diversos métodos depasivación tales como el crecimiento de las capas intermedias (Si, Ge y Si/Ge) y/otratamientos químicos para mejorar la calidad de la interfaz, fenómenos como ladifusión de las especies atómicas del sustrato como consecuencia del recocido no hansido cuidadosamente estudiados. Los tratamientos térmicos, los cuales estánrelacionados con los procesos de integración de la fuente y el drenador (S/D) en undispositivo MOSFET, conducen a cambios estructurales que degradan el rendimientoeléctrico de un dispositivo MOS. Una caracterización apropiada de las alteracionesestructurales asociadas con la difusión de los elementos del substrato hacia las capassuperiores es importante para entender cuáles son los mecanismos de falla en undispositivo MOS. En este trabajo se presenta un análisis de la estructura y laestabilidad térmica de la estructura TiN/ZrO2/InGaAs por la espectroscopía defotoelectrones por rayos X con resolución angular (ARXPS). A través de un método deanálisis no destructivo, fue posible observar efectos sutiles tales como la difusión delas especies atómicas del sustrato a través del dieléctrico como consecuencia delrecocido. El conocimiento detallado de la estructura permitió evaluar los perfiles deprofundidad para las componentes de In-O y Ga-O por medio del método deescenarios. Para la muestra en estado como se depositó, las componentes de In-O yGa-O fueron localizadas en la interfaz óxido-semiconductor. Después del recocido, semuestra cuantitativamente que éste causa la difusión de átomos de In y Ga hacia a lascapas superiores. Asumiendo diferentes escenarios para su distribución, se muestraque el recocido provoca la difusión de In y Ga hasta la capa de TiO2. Para la muestrarecocida a 500 °C, se observó claramente la difusión de indio, mientras que para lamuestra recocida a 700 °C tanto In y Ga difunden a la capa de TiO2. El análisis~ iv ~cuantitativo mostró que existe menor difusión de átomos de galio (0.12 ML) que deindio (0.26 ML) a 700 °C/10 s. Puesto que el análisis sobre la cantidad de materialdifundido se realizó a diferentes temperaturas, fue posible obtener un valoraproximado para la energía de activación del indio a través del ZrO2. El valor resultóser muy similar a los valores reportados previamente para la difusión de indio a travésde Al2O3 y a través de HfO2. Con el fin de correlacionar los resultados obtenidos porARXPS, se emplearon técnicas complementarias como la microscopía electrónica detransmisión (TEM), la espectroscopía de energía dispersiva (EDX) y la espectrometríade masas de iones secundarios por tiempo de vuelo (SIMS-TOF). Particularmente, TOFSIMSdestacó el fenómeno de difusión de las especies atómicas sustrato hacia lasuperficie
- …
