Compared to traditional compile-time optimization, run-time optimization could offer significant performance improvements when parallelizing and optimizing adaptive irregular applications, because it performs program analysis and adaptive optimizations during program execution. Run-time techniques can succeed where static techniques fail because they exploit the characteristics of input data, programs' dynamic behaviors, and the underneath execution environment. When optimizing adaptive irregular applications for parallel execution, a common observation is that the effectiveness of the optimizing transformations depends on programs' input data and their dynamic phases. This dissertation presents a set of run-time optimization techniques that match the characteristics of programs' dynamic memory access patterns and the appropriate optimization (parallelization) transformations. First, we present a general adaptive algorithm selection framework to automatically and adaptively select at run-time the best performing, functionally equivalent algorithm for each of its execution instances. The selection process is based on off-line automatically generated prediction models and characteristics (collected and analyzed dynamically) of the algorithm's input data, In this dissertation, we specialize this framework for automatic selection of reduction algorithms. In this research, we have identified a small set of machine independent high-level characterization parameters and then we deployed an off-line, systematic experiment process to generate prediction models. These models, in turn, match the parameters to the best optimization transformations for a given machine. The technique has been evaluated thoroughly in terms of applications, platforms, and programs' dynamic behaviors. Specifically, for the reduction algorithm selection, the selected performance is within 2% of optimal performance and on average is 60% better than "Replicated Buffer," the default parallel reduction algorithm specified by OpenMP standard. To reduce the overhead of speculative run-time parallelization, we have developed an adaptive run-time parallelization technique that dynamically chooses effcient shadow structures to record a program's dynamic memory access patterns for parallelization. This technique complements the original speculative run-time parallelization technique, the LRPD test, in parallelizing loops with sparse memory accesses. The techniques presented in this dissertation have been implemented in an optimizing research compiler and can be viewed as effective building blocks for comprehensive run-time optimization systems, e.g., feedback-directed optimization systems and dynamic compilation systems.

Compared to traditional compile-time optimization, run-time optimization could offer significant performance improvements when parallelizing and optimizing adaptive irregular applications, because it performs program analysis and adaptive optimizations during program execution. Run-time techniques can succeed where static techniques fail because they exploit the characteristics of input data, programs' dynamic behaviors, and the underneath execution environment. When optimizing adaptive irregular applications for parallel execution, a common observation is that the effectiveness of the optimizing transformations depends on programs' input data and their dynamic phases. This dissertation presents a set of run-time optimization techniques that match the characteristics of programs' dynamic memory access patterns and the appropriate optimization (parallelization) transformations. First, we present a general adaptive algorithm selection framework to automatically and adaptively select at run-time the best performing, functionally equivalent algorithm for each of its execution instances. The selection process is based on off-line automatically generated prediction models and characteristics (collected and analyzed dynamically) of the algorithm's input data, In this dissertation, we specialize this framework for automatic selection of reduction algorithms. In this research, we have identified a small set of machine independent high-level characterization parameters and then we deployed an off-line, systematic experiment process to generate prediction models. These models, in turn, match the parameters to the best optimization transformations for a given machine. The technique has been evaluated thoroughly in terms of applications, platforms, and programs' dynamic behaviors. Specifically, for the reduction algorithm selection, the selected performance is within 2% of optimal performance and on average is 60% better than "Replicated Buffer," the default parallel reduction algorithm specified by OpenMP standard. To reduce the overhead of speculative run-time parallelization, we have developed an adaptive run-time parallelization technique that dynamically chooses effcient shadow structures to record a program's dynamic memory access patterns for parallelization. This technique complements the original speculative run-time parallelization technique, the LRPD test, in parallelizing loops with sparse memory accesses. The techniques presented in this dissertation have been implemented in an optimizing research compiler and can be viewed as effective building blocks for comprehensive run-time optimization systems, e.g., feedback-directed optimization systems and dynamic compilation systems.

Les structures assemblées par points de soudure doivent satisfaire à un grand nombre de critères de performance, portant entre autres sur des aspects statiques, dynamiques, acoustiques, ou crash, qui sont fortement influencés par le nombre et la position des points de soudure. De plus, certains d’entre eux peuvent être défaillants en sortie de ligne d’assemblage ou le devenir au cours de la vie du véhicule, ce qui induit des dispersions des performances. Le challenge est donc de déterminer, dans un temps d’analyse raisonnable, le nombre et la répartition optimale des points de soudure permettant la vérification des objectifs de performance tout en garantissant le meilleur compromis entre le coût de fabrication et la robustesse des performances. Une procédure adaptative est proposée, permettant à chaque itération d’ajouter et de supprimer des points de soudure, afin d’obtenir une conception robuste. Une procédure itérative de robustesse info-gap est également présentée, afin d’analyser l’impact de la perte des points de soudure les plus influents. On définit ainsi une procédure de contrôle qualité, portant sur un petit nombre de points de soudure, permettant de garantir l’homogénéité des performances des structures fabriquées en série. Toutes les méthodologies sont validées sur une caisse en blanc de véhicule industriel.En pratique, la conception doit satisfaire plusieurs critères antagonistes. La procédure adaptative est donc étendue à l’optimisation multi-objectifs, afin d’obtenir le meilleur compromis possible en fonction des critères considérés. Une illustration portant sur l’optimisation dynamique et vibro-acoustique d’une cavité passager est finalement mené / Spot welded structures must satisfy a large variety of performance characteristics including static, noise, vibration and harshness and crash, which are strongly influenced by the numbers and positions of the spot welds. Moreover, some spot welds may be missing when structures leave the assembly line or break during lifetime of vehicle, inducing large variations in the performance characteristics. The challenge is to find the best locations and number of spot welds to achieve the performance goals while taking into account the trade-offs between manufacturing cost and robust to uncertainty in reasonable time frame. An adaptive optimization procedure is proposed, which iteratively adds and removes spot welds to obtain the robust design efficiently. An iterative procedure is also proposed to obtain the info-gap robustness curve, which allows analyzing the impact of the loss of the most influential spot welds on performance. This leads to define a quality control procedure for a small number of spot welds to guarantee the performance of a population of nominally identical structures within acceptable limits. All the methodologies are validated on a full industrial body-in-white of a vehicle.In practice, the design must satisfy several stringent and conflicting objectives simultaneously. Thus, we extend the adaptive procedure to handle multi-objective problems using the weight-sum approach reflecting the importance given to individual spot welds to obtain efficient solutions in the objective space. Finally, we propose a strategy to optimize the vibro-acoustic behavior of the passenger cabin of a vehicle by designing the numbers and positions of the spot welds.

Les représentations convolutives extraient des motifs récurrents qui aident à comprendre la structure locale dans un jeu de signaux. Elles sont adaptées pour l’analyse des signaux physiologiques, qui nécessite des visualisations mettant en avant les informations pertinentes. Ces représentations sont aussi liées aux modèles d’apprentissage profond. Dans ce manuscrit, nous décrivons des avancées algorithmiques et théoriques autour de ces modèles. Nous montrons d’abord que l’Analyse du Spectre Singulier permet de calculer efficacement une représentation convolutive. Cette représentation est dense et nous décrivons une procédure automatisée pour la rendre plus interprétable. Nous proposons ensuite un algorithme asynchrone, pour accélérer le codage parcimonieux convolutif. Notre algorithme présente une accélération super-linéaire. Dans une seconde partie, nous analysons les liens entre représentations et réseaux de neurones. Nous proposons une étape d’apprentissage supplémentaire, appelée post-entraînement, qui permet d’améliorer les performances du réseau entraîné, en s’assurant que la dernière couche soit optimale. Puis nous étudions les mécanismes qui rendent possible l’accélération du codage parcimonieux avec des réseaux de neurones. Nous montrons que cela est lié à une factorisation de la matrice de Gram du dictionnaire. Finalement, nous illustrons l’intérêt de l’utilisation des représentations convolutives pour les signaux physiologiques. L’apprentissage de dictionnaire convolutif est utilisé pour résumer des signaux de marche et le mouvement du regard est soustrait de signaux oculométriques avec l’Analyse du Spectre Singulier. / Convolutional representations extract recurrent patterns which lead to the discovery of local structures in a set of signals. They are well suited to analyze physiological signals which requires interpretable representations in order to understand the relevant information. Moreover, these representations can be linked to deep learning models, as a way to bring interpretability intheir internal representations. In this disserta tion, we describe recent advances on both computational and theoretical aspects of these models.First, we show that the Singular Spectrum Analysis can be used to compute convolutional representations. This representation is dense and we describe an automatized procedure to improve its interpretability. Also, we propose an asynchronous algorithm, called DICOD, based on greedy coordinate descent, to solve convolutional sparse coding for long signals. Our algorithm has super-linear acceleration.In a second part, we focus on the link between representations and neural networks. An extra training step for deep learning, called post-training, is introduced to boost the performances of the trained network by making sure the last layer is optimal. Then, we study the mechanisms which allow to accelerate sparse coding algorithms with neural networks. We show that it is linked to afactorization of the Gram matrix of the dictionary.Finally, we illustrate the relevance of convolutional representations for physiological signals. Convolutional dictionary learning is used to summarize human walk signals and Singular Spectrum Analysis is used to remove the gaze movement in young infant’s oculometric recordings.

The increasing size and complexity of datasets have accelerated the development of machine learning models and exposed the need for more scalable solutions. This thesis explores challenges associated with large-scale machine learning under data privacy constraints. With the growth of machine learning models, traditional privacy methods such as data anonymization are becoming insufficient. Thus, we delve into alternative approaches, such as differential privacy. Our research addresses the following core areas in the context of scalable privacy-preserving machine learning: First, we examine the implications of data dimensionality on privacy for the application of medical image analysis. We extend the classification algorithm Private Aggregation of Teacher Ensembles (PATE) to deal with high-dimensional labels, and demonstrate that dimensionality reduction can be used to improve privacy. Second, we consider the impact of hyperparameter selection on privacy. Here, we propose a novel adaptive technique for hyperparameter selection in differentially gradient-based optimization. Third, we investigate sampling-based solutions to scale differentially private machine learning to dataset with a large number of records. We study the privacy-enhancing properties of importance sampling, highlighting that it can outperform uniform sub-sampling not only in terms of sample efficiency but also in terms of privacy. The three techniques developed in this thesis improve the scalability of machine learning while ensuring robust privacy protection, and aim to offer solutions for the effective and safe application of machine learning in large datasets. / Den ständigt ökande storleken och komplexiteten hos datamängder har accelererat utvecklingen av maskininlärningsmodeller och gjort behovet av mer skalbara lösningar alltmer uppenbart. Den här avhandlingen utforskar tre utmaningar förknippade med storskalig maskininlärning under dataskyddskrav. För stora och komplexa maskininlärningsmodeller blir traditionella metoder för integritet, såsom datananonymisering, otillräckliga. Vi undersöker därför alternativa tillvägagångssätt, såsom differentiell integritet. Vår forskning behandlar följande utmaningar inom skalbar och integitetsmedveten maskininlärning: För det första undersöker vi hur hög data-dimensionalitet påverkar integriteten för medicinsk bildanalys. Vi utvidgar klassificeringsalgoritmen Private Aggregation of Teacher Ensembles (PATE) för att hantera högdimensionella etiketter och visar att dimensionsreducering kan användas för att förbättra integriteten. För det andra studerar vi hur valet av hyperparametrar påverkar integriteten. Här föreslår vi en ny adaptiv teknik för val av hyperparametrar i gradient-baserad optimering med garantier på differentiell integritet. För det tredje granskar vi urvalsbaserade lösningar för att skala differentiellt privat maskininlärning till stora datamängder. Vi studerar de integritetsförstärkande egenskaperna hos importance sampling och visar att det kan överträffa ett likformigt urval av sampel, inte bara när det gäller effektivitet utan även för integritet. De tre teknikerna som utvecklats i denna avhandling förbättrar skalbarheten för integritetsskyddad maskininlärning och syftar till att erbjuda lösningar för effektiv och säker tillämpning av maskininlärning på stora datamängder. / <p>QC 20231101</p>