A big challenge in artificial intelligence (AI) is creating autonomous agents that can interact well with other agents over extended periods of time. Most previously developed algorithms have been designed in the context of Repeated Games, environments in which the agents interact in the same scenario repeatedly. However, in most real-world interactions, relationships between people and autonomous agents consist of sequences of distinct encounters with different incentives and payoff structures. Therefore, in this thesis, we consider Interaction Games, which model interactions in which the scenario changes from encounter to encounter, often in ways that are unanticipated by the players. For example, in Interaction Games, the magnitude of payoffs as well as the structure of these payoffs can differ across encounters. Unfortunately, while there have been many algorithms developed for Repeated Games, there are no known algorithms for playing Interaction Games. Thus, we have developed two different algorithms, augmented Fictitious Play (aFP) and augmented S# (Aug-S#), for playing these games. These algorithms are designed to generalize Fictitious Play and S# algorithms, which were previously created for Repeated Games, to the more general kinds of scenarios modeled by Interaction Games. This thesis primarily focuses on the evaluation of these algorithms. We first analyze the behavioral and performance properties of these algorithms when associating with other autonomous algorithms. We then report on the results of a user study in which these algorithms were paired with people in two different Interaction Games. Our results show that while the generalized algorithms demonstrate many of the same properties in Interaction Games as they do in Repeated Games, the complexity of Interaction Games appear to alter the kinds of behaviors that are successful, particularly in environments in which communication between players is not possible.

Machine Learning is growing in popularity and spreading across different fields for various applications. Due to this trend, machine learning algorithms use different hardware platforms and are being experimented to obtain high test accuracy and throughput. FPGAs are well-suited hardware platform for machine learning because of its re-programmability and lower power consumption. Programming using FPGAs for machine learning algorithms requires substantial engineering time and effort compared to software implementation. We propose a software assisted design flow to program FPGA for machine learning algorithms using our hardware library. The hardware library is highly parameterized and it accommodates Tree Classifiers. As of now, our library consists of the components required to implement decision trees and random forests. The whole automation is wrapped around using a python script which takes you from the first step of having a dataset and design choices to the last step of having a hardware descriptive code for the trained machine learning model.

Cette thèse porte sur la configurationAutomatisée des algorithmes qui vise à trouver le meilleur paramétrage à un problème donné ou une catégorie deproblèmes.Le problème de configuration de l'algorithme revient doncà un problème de métaFoptimisation dans l'espace desparamètres, dont le métaFobjectif est la mesure deperformance de l’algorithme donné avec une configuration de paramètres donnée.Des approches plus récentes reposent sur une description des problèmes et ont pour but d’apprendre la relationentre l’espace des caractéristiques des problèmes etl’espace des configurations de l’algorithme à paramétrer.Cette thèse de doctorat porter le CAPI (Configurationd'Algorithme Par Instance) pour résoudre des problèmesd'optimisation de boîte noire continus, où seul un budgetlimité d'évaluations de fonctions est disponible. Nous étudions d'abord' les algorithmes évolutionnairesPour l'optimisation continue, en mettant l'accent sur deux algorithmes que nous avons utilisés comme algorithmecible pour CAPI,DE et CMAFES.Ensuite, nous passons en revue l'état de l'art desapproches de configuration d'algorithme, et lesdifférentes fonctionnalités qui ont été proposées dansla littérature pour décrire les problèmesd'optimisation de boîte noire continue.Nous introduisons ensuite une méthodologie générale Pour étudier empiriquement le CAPI pour le domainecontinu, de sorte que toutes les composantes du CAPIpuissent être explorées dans des conditions réelles.À cette fin, nous introduisons également un nouveau Banc d'essai de boîte noire continue, distinct ducélèbre benchmark BBOB, qui est composé deplusieurs fonctions de test multidimensionnelles avec'différentes propriétés problématiques, issues de lalittérature.La méthodologie proposée est finalement appliquée 'àdeux AES. La méthodologie est ainsi, validéempiriquement sur le nouveau banc d’essaid’optimisation boîte noire pour des dimensions allant jusqu’à 100. / This PhD thesis focuses on the automated algorithm configuration that aims at finding the best parameter setting for a given problem or a' class of problem. The Algorithm Configuration problem thus amounts to a metal Foptimization problem in the space of parameters, whosemetaFobjective is the performance measure of the given algorithm at hand with a given parameter configuration. However, in the continuous domain, such method can only be empirically assessed at the cost of running the algorithm on some problem instances. More recent approaches rely on a description of problems in some features space, and try to learn a mapping from this feature space onto the space of parameter configurations of the algorithm at hand. Along these lines, this PhD thesis focuses on the Per Instance Algorithm Configuration (PIAC) for solving continuous black boxoptimization problems, where only a limited budget confessionnalisations available. We first survey Evolutionary Algorithms for continuous optimization, with a focus on two algorithms that we have used as target algorithm for PIAC, DE and CMAFES. Next, we review the state of the art of Algorithm Configuration approaches, and the different features that have been proposed in the literature to describe continuous black box optimization problems. We then introduce a general methodology to empirically study PIAC for the continuous domain, so that all the components of PIAC can be explored in real Fworld conditions. To this end, we also introduce a new continuous black box test bench, distinct from the famous BBOB'benchmark, that is composed of a several multiFdimensional test functions with different problem properties, gathered from the literature. The methodology is finally applied to two EAS. First we use Differential Evolution as'target algorithm, and explore all the components of PIAC, such that we empirically assess the best. Second, based on the results on DE, we empirically investigate PIAC with Covariance Matrix Adaptation Evolution Strategy (CMAFES) as target algorithm. Both use cases empirically validate the proposed methodology on the new black box testbench for dimensions up to100.

Le Transport Optimal régularisé par l’Entropie (TOE) permet de définir les Divergences de Sinkhorn (DS), une nouvelle classe de distance entre mesures de probabilités basées sur le TOE. Celles-ci permettentd’interpolerentredeuxautresdistancesconnues: leTransport Optimal(TO)etl’EcartMoyenMaximal(EMM).LesDSpeuventêtre utilisées pour apprendre des modèles probabilistes avec de meilleures performances que les algorithmes existants pour une régularisation adéquate. Ceci est justifié par un théorème sur l’approximation des SDpardeséchantillons, prouvantqu’unerégularisationsusantepermet de se débarrasser de la malédiction de la dimension du TO, et l’on retrouve à l’infini le taux de convergence des EMM. Enfin, nous présentons de nouveaux algorithmes de résolution pour le TOE basés surl’optimisationstochastique‘en-ligne’qui,contrairementàl’étatde l’art, ne se restreignent pas aux mesures discrètes et s’adaptent bien aux problèmes de grande dimension. / This thesis proposes theoretical and numerical contributions to use Entropy-regularized Optimal Transport (EOT) for machine learning. We introduce Sinkhorn Divergences (SD), a class of discrepancies betweenprobabilitymeasuresbasedonEOTwhichinterpolatesbetween two other well-known discrepancies: Optimal Transport (OT) and Maximum Mean Discrepancies (MMD). We develop an ecient numerical method to use SD for density fitting tasks, showing that a suitable choice of regularization can improve performance over existing methods. We derive a sample complexity theorem for SD which proves that choosing a large enough regularization parameter allows to break the curse of dimensionality from OT, and recover asymptotic ratessimilartoMMD.Weproposeandanalyzestochasticoptimization solvers for EOT, which yield online methods that can cope with arbitrary measures and are well suited to large scale problems, contrarily to existing discrete batch solvers.

The ability to diagnose deviations and predict faults effectively is an important task in various industrial domains for minimizing costs and productivity loss and also conserving environmental resources. However, the majority of the efforts for diagnostics are still carried out by human experts in a time-consuming and expensive manner. Automated data-driven solutions are needed for continuous monitoring of complex systems over time. On the other hand, domain expertise plays a significant role in developing, evaluating, and improving diagnostics and monitoring functions. Therefore, automatically derived solutions must be able to interact with domain experts by taking advantage of available a priori knowledge and by incorporating their feedback into the learning process. This thesis and appended papers tackle the problem of generating a real-world self-monitoring system for continuous monitoring of machines and operations by developing algorithms that can learn data streams and their relations over time and detect anomalies using joint-human machine learning. Throughout this thesis, we have described a number of different approaches, each designed for the needs of a self-monitoring system, and have composed these methods into a coherent framework. More specifically, we presented a two-layer meta-framework, in which the first layer was concerned with learning appropriate data representations and detectinganomalies in an unsupervised fashion, and the second layer aimed at interactively exploiting available expert knowledge in a joint human-machine learning fashion. Furthermore, district heating has been the focus of this thesis as the application domain with the goal of automatically detecting faults and anomalies by comparing heat demands among different groups of customers. We applied and enriched different methods on this domain, which then contributed to the development and improvement of the meta-framework. The contributions that result from the studies included in this work can be summarized into four categories: (1) exploring different data representations that are suitable for the self-monitoring task based on data characteristics and domain knowledge, (2) discovering patterns and groups in data that describe normal behavior of the monitored system/systems, (3) implementing methods to successfully discriminate anomalies from the normal behavior, and (4) incorporating domain knowledge and expert feedback into self-monitoring.

Biometrics is the study of physical or behavioral traits that establishes the identity of a person. Forensics, physical security and cyber security are some of the main fields that use biometrics. Unlike traditional authentication systems—such as password based—biometrics cannot be lost, forgotten or shared. This is possible because biometrics establishes the identity of a person based on a physiological/behavioural characteristic rather than what the person possess or remembers. Biometrics has two modes of operation: identification and authentication. Identification finds the identity of a person among a group of persons. Authentication determines if the claimed identity of a person is truthful. Biometric person authentication is an alternative to passwords or graphical patterns. It prevents shoulder surfing attacks, i.e., people watching from a short distance. Nevertheless, biometric traits of conventional authentication techniques like fingerprints, face—and to some extend iris—are easy to capture and duplicate. This denotes a security risk for modern and future applications such as digital twins, where an attacker can copy and duplicate a biometric trait in order to spoof a biometric system. Researchers have proposed ECG as biometric authentication to solve this problem. ECG authentication conceals the biometric traits and reduces the risk of an attack by duplication of the biometric trait. However, current ECG authentication solutions require 10 or more seconds of an ECG signal in order to have accurate results. The accuracy is directly proportional to the ECG signal time-length for authentication. This is inconvenient to implement ECG authentication in an end-user product because a user cannot wait 10 or more seconds to gain access in a secure manner to their device. This thesis addresses the problem of spoofing by proposing an accurate and secure ECG biometric authentication system with relatively short ECG signal length for authentication. The system consists of an ECG acquisition from lead I (two electrodes), signal processing approaches for filtration and R-peak detection, a feature extractor and an authentication process. To evaluate this system, we developed a method to calculate the Equal Error Rate—EER—with non-normal distributed data. In the authentication process, we propose an approach based on Support Vector Machine—SVM—and achieve 4.5% EER with 4 seconds of ECG signal length for authentication. This approach opens the door for a deeper understanding of the signal and hence we enhanced it by applying a hybrid approach of Convolutional Neural Networks—CNN—combined with SVM. The purpose of this hybrid approach is to improve accuracy by automatically detect and extract features with Deep Learning—in this case CNN—and then take the output into a one-class SVM classifier—Authentication; which proved to outperform accuracy for one-class ECG classification. This hybrid approach reduces the EER to 2.84% with 4 seconds of ECG signal length for authentication. Furthermore, we investigated the combination of two different biometrics techniques and we improved the accuracy to 0.46% EER, while maintaining a short ECG signal length for authentication of 4 seconds. We fuse Fingerprint with ECG at the decision level. Decision level fusion requires information that is available from any biometric technique. Fusion at different levels—such as feature level fusion—requires information about features that are incompatible or hidden. Fingerprint minutiae are composed of information that differs from ECG peaks and valleys. Therefore fusion at the feature level is not possible unless the fusion algorithm provides a compatible conversion scheme. Proprietary biometric hardware does not provide information about the features or the algorithms; therefore, features are hidden and not accessible for feature level fusion; however, the result is always available for a decision level fusion.

This dissertation describes a study which aims to classify video flows from Internet network traffic. In this study, classification is done based on the characteristics of the flow, which includes features such as payload sizes and inter-arrival time. The purpose of this is to give an alternative to classifying flows based on the contents of their payload packets. Because of an increase of encrypted flows within Internet network traffic, this is a necessity. Data with known class is fed to a machine learning classifier such that a model can be created. This model can then be used for classification of new unknown data. For this study, two different classifiers are used, namely decision trees and random forest. Several tests are completed to attain the best possible models. The results of this dissertation shows that classification based on characteristics is possible and the random forest classifier in particular achieves good accuracies. However, the accuracy of classification of encrypted flows was not able to be tested within this project. / HITS, 4707

Thesis advisor: Ziqiang . Wang / The transition between many-body localized states and the delocalized thermal states is an eigenstate phase transition at finite energy density outside the scope of conventional quantum statistical mechanics. We apply support vector machine (SVM) to study the phase transition between many-body localized and thermal phases in a disordered quantum Ising chain in a transverse external field. The many-body eigenstate energy E is bounded by a bandwidth W=Eₘₐₓ-Eₘᵢₙ. The transition takes place on a phase diagram spanned by the energy density ϵ=2(Eₘₐₓ-Eₘᵢₙ)/W and the disorder strength ẟJ of the spin interaction uniformly distributed within [-ẟJ, ẟJ], formally parallel to the mobility edge in Anderson localization. In our study we use the labeled probability density of eigenstate wavefunctions belonging to the deeply localized and thermal regimes at two different energy densities (ϵ's) as the training set, i.e., providing labeled data at four corners of the phase diagram. Then we employ the trained SVM to predict the whole phase diagram. The obtained phase boundary qualitatively agrees with previous work using entanglement entropy to characterize these two phases. We further analyze the decision function of the SVM to interpret its physical meaning and find that it is analogous to the inverse participation ratio in configuration space. Our findings demonstrate the ability of the SVM to capture potential quantities that may characterize the many-body localization phase transition. To further investigate the properties of the transition, we study the behavior of the entanglement entropy of a subsystem of size L_A in a system of size L > L_A near the critical regime of the many-body localization transition. The many-body eigenstates are obtained by exact diagonalization of a disordered quantum spin chain under twisted boundary conditions to reduce the finite-size effect. We present a scaling theory based on the assumption that the transition is continuous and use the subsystem size L_A/ξ as the scaling variable, where ξ is the correlation length. We show that this scaling theory provides an effective description of the critical behavior and that the entanglement entropy follows the thermal volume law at the transition point. We extract the critical exponent governing the divergence of ξ upon approaching the transition point. We again study the participation entropy in the spin-basis of the domain wall excitations and show that the transition point and the critical exponent agree with those obtained from finite size scaling of the entanglement entropy. Our findings suggest that the many-body localization transition in this model is continuous and describable as a localization transition in the many-body configuration space. Besides the many-body localization transition driven by disorder, We also study the Coulomb repulsion and temperature driving phase transitions. We apply a finite-temperature Gutzwiller projection to two-dimensional Hubbard model by constructing a "Gutzwiller-type" density matrix operator to approximate the real interacting density matrix, which provides the upper bound of free energy of the system. We firstly investigate half filled Hubbard model without magnetism and obtain the phase diagram. The transition line is of first order at finite temperature, ending at 2 second order points, which shares qualitative agreement with dynamic mean field results. We derive the analytic form of the free energy and therefor the equation of states, which benefits the understanding of the different phases. We later extend our approach to take anti-ferromagnetic order into account. We determine the Neel temperature and explore its interesting behavior when varying the Coulomb repulsion. / Thesis (PhD) — Boston College, 2019. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Physics.

As machine learning becomes an increasingly relevant field being incorporated into everyday life, so does the need for consistently high performing models. With these high expectations, along with potentially restrictive data sets, it is crucial to be able to use techniques for machine learning that increase the likelihood of success. Robust Principal Component Analysis (RPCA) not only extracts anomalous data, but also finds correlations among the given features in a data set, in which these correlations can themselves be used as features. By taking a novel approach to utilizing the output from RPCA, we address how our method effects the performance of such models. We take into account the efficiency of our approach, and use projectors to enable our method to have a 99.79% faster run time. We apply our method primarily to cyber security data sets, though we also investigate the effects on data sets from other fields (e.g. medical).

As online casino and betting increases in popularity across the globe, the importance of green gambling has become an important subject of discussion. The Swedish betting company, ATG, realises the benefits of this and would like to prevent their gamblers from falling into problem gambling. To predict problem gambling, Bayesian networks were trained on previously identified problem gamblers, separated into seven risk groups. The network was then able to predict the risk group of previously unseen gamblers with an ac- curacy of 94%. It also achieved an average precision of 89%, an average recall of 96% and an average f1-score of 93%. The features in the data set were also ranked, to find which were most important in predicting problem gambling. It was found that municipality, which day of the week the transaction was made and during which hour of the day were the most important features. Also, the Bayesian network was also made as simple as possible, by removing irrelevant features and features which carry very low importance.