This work presents an investigation of tailoring Network Representation Learning (NRL) for an application in the Financial Industry. NRL approaches are data-driven models that learn how to encode graph structures into low-dimensional vector spaces, which can be further exploited by downstream Machine Learning applications. They can potentially bring a lot of benefits in the Financial Industry since they extract in an automatic way features that can provide useful input regarding graph structures, called embeddings. Financial transactions can be represented as a network, and through NRL, it is possible to extract embeddings that reflect the intrinsic inter-connected nature of economic relationships. Such embeddings can be used for several purposes, among which Anomaly Detection to fight financial crime.This work provides a qualitative analysis over state-of-the-art NRL models, which identifies Graph Convolutional Network (ConvGNN) as the most suitable category of approaches for Financial Industry but with a certain need for further improvement. Financial Industry poses additional challenges when modelling a NRL solution. Despite the need of having a scalable solution to handle real-world graph with considerable dimensions, it is necessary to take into consideration several characteristics: transactions graphs are inherently dynamic since every day new transactions are executed and nodes can be heterogeneous. Besides, everything is further complicated by the need to have updated information in (near) real-time due to the sensitivity of the application domain. For these reasons, GraphSAGE has been considered as a base for the experiments, which is an inductive ConvGNN model. Two variants of GraphSAGE are presented: a dynamic variant whose weights evolve accordingly with the input sequence of graph snapshots, and a variant specifically meant to handle bipartite graphs. These variants have been evaluated by applying them to real-world data and leveraging the generated embeddings to perform Anomaly Detection. The experiments demonstrate that leveraging these variants leads toimagecomparable results with other state-of-the-art approaches, but having the advantage of being suitable to handle real-world financial data sets. / Detta arbete presenterar en undersökning av tillämpningar av Network Representation Learning (NRL) inom den finansiella industrin. Metoder inom NRL möjliggör datadriven kondensering av grafstrukturer till lågdimensionella och lätthanterliga vektorer.Dessa vektorer kan sedan användas i andra maskininlärningsuppgifter. Närmare bestämt, kan metoder inom NRL underlätta hantering av och informantionsutvinning ur beräkningsintensiva och storskaliga grafer inom den finansiella sektorn, till exempel avvikelsehantering bland finansiella transaktioner. Arbetet med data av denna typ försvåras av det faktum att transaktionsgrafer är dynamiska och i konstant förändring. Utöver detta kan noderna, dvs transaktionspunkterna, vara vitt skilda eller med andra ord härstamma från olika fördelningar.I detta arbete har Graph Convolutional Network (ConvGNN) ansetts till den mest lämpliga lösningen för nämnda tillämpningar riktade mot upptäckt av avvikelser i transaktioner. GraphSAGE har använts som utgångspunkt för experimenten i två olika varianter: en dynamisk version där vikterna uppdateras allteftersom nya transaktionssekvenser matas in, och en variant avsedd särskilt för bipartita (tvådelade) grafer. Dessa varianter har utvärderats genom användning av faktiska datamängder med avvikelsehantering som slutmål.

Multi-object tracking is a classic engineering problem wherein a system must keep track of the identities of a set of a priori unknown objects through a sequence, for example video. Perfect execution of this task would mean no spurious or missed detections or identities, neither swapped identities. To measure performance of tracking systems, the Higher Order Tracking Accuracy metric is often used, which takes into account both detection and association accuracy. Prior work in monocular vision-based multi-object tracking has integrated deep learning to various degrees, with deep learning based detectors and visual feature extractors being commonplace alongside motion models of varying complexities. These methods have historically combined the usage of position and appearance in their association stage using hand-crafted heuristics, featuring increasingly complex algorithms to achieve higher performance tracking. With an interest in simplifying tracking algorithms, we turn to the field of representation learning. Presenting a novel method using a Joint-Embedding Predictive Architecture, trained through a contrastive objective, we learn object feature embeddings initialized by detections from a pre-trained detector. The results are features that fuse both positional and visual features. Comparing the performance of our method on the complex DanceTrack and relatively simpler MOT17 datasets to that of the most performant heuristic-based alternative, Deep OC-SORT, we see a significant improvement of 66.1 HOTA compared to the 61.3 HOTA of Deep OC-SORT on DanceTrack. On MOT17, which features less complex motion and less training data, heuristics-based methods outperform the proposed and prior learned tracking methods. While the method lags behind the state of the art in complex scenes, which follows the tracking-by-attention paradigm, it presents a novel approach and brings with it a new avenue of possible research. / Spårning av multipla objekt är ett typiskt ingenjörsproblem där ett system måste hålla reda på identiteterna hos en uppsättning på förhand okända objekt genom en sekvens, till exempel video. Att perfekt utföra denna uppgift skulle innebära inga felaktiga eller missade detektioner eller identiteter, inte heller utbytta identiteter. För att mäta prestanda hos spårningssystem används ofta metriken HOTA, som tar hänsyn till både detektions- och associationsnoggrannhet. Tidigare arbete inom monokulär vision-baserad flerobjektsspårning har integrerat djupinlärning i olika grad, med detektorer baserade på djupinlärning och visuella funktionsutdragare som är vanliga tillsammans med rörelsemodeller av varierande komplexitet. Dessa metoder har historiskt kombinerat användningen av position och utseende i deras associationsfas med hjälp av handgjorda heuristiker, med alltmer komplexa algoritmer för att uppnå högre prestanda i spårningen. Med ett intresse för att förenkla spårningsalgoritmer, vänder vi oss till fältet för representationsinlärning. Vi presenterar en ny metod som använder en prediktiv arkitektur med gemensam inbäddning, tränad genom ett kontrastivt mål, där vi lär oss objekt representationer initierade av detektioner från en förtränad detektor. Resultatet är en funktion som sammansmälter både position och visuel information. När vi jämför vår metod på det komplexa DanceTrack och det relativt enklare MOT17-datasetet med det mest presterande heuristikbaserade alternativet, Deep OC-SORT, ser vi en betydande förbättring på 66,1 HOTA jämfört med 61,3 HOTA för Deep OC-SORT på DanceTrack. På MOT17, som har mindre komplex rörelse och mindre träningsdata, presterar heuristikbaserade metoder bättre än den föreslagna och tidigare lärande spårningsmetoderna. Även om metoden ligger efter den senaste utvecklingen i komplexa scener, som följer paradigm för spårning-genom-uppmärksamhet, presenterar den ett nytt tillvägagångssätt och för med sig möjligheter för ny forskning.

<p dir="ltr">This dissertation introduces a novel graph embedding paradigm, leveraging a force-directed scheme for graph embedding. In the field of graph embedding, an "embedding" refers to the process of transforming elements of a graph such as nodes, or edges, or potentially other structural information of the graph into a low-dimensional space, typically a vector space, while preserving the graph's structural properties as much as possible. The dimensions of the space are supposed to be much smaller than the elements of the graph that are to be embedded. This transformation results in a set of vectors, with each vector representing a node (or edge) in the graph. The goal is to capture the essence of the graph's topology, node connectivity, and other relevant features in a way that facilitates easier processing by machine learning algorithms, which often perform better with input data in a continuous vector space.</p><p dir="ltr">The main premise of kinematics-based force-directed graph embedding is that the nodes are considered as massive mobile objects that can be moved around in the embedding space under force. In this PhD thesis, we devised a general theoretical framework for the proposed graph embedding paradigm and provided the mathematical proof of convergence given the required constraints. From this point on, the objective was to explore force functions and parameters and methods of applying them in terms of their efficacy regarding graph embedding applications. We found some force functions that outperformed the state-of-the-art methods.</p><p dir="ltr">The author of this manuscript believes that the proposed paradigm will open a new chapter, specifically in the field of graph embedding and generally in the field of embedding.</p>

<p dir="ltr">Graph-structured data capture complex relationships and interactions between entities, offering valuable insights for scientific discovery, business modeling, and AI-driven decision-making. Despite its transformative potential, learning on graphs faces two key challenges: (1) scaling expressive learning approaches, especially subgraph-based graph representation learning, and (2) ensuring privacy when handling sensitive relational data. Both challenges arise from intricate dependencies in graph structures, which limit the effectiveness of canonical algorithms and system optimizations. This dissertation addresses these challenges through a unified framework that integrates system-aware algorithm design across two main thrusts.</p><p dir="ltr">In Thrust I, we develop a family of efficient frameworks for expressive graph representation learning that eliminate redundancy in subgraph-based methods. By decoupling dependencies over task-specific input features (i.e., query-induced subgraphs), the proposed paradigm enables efficient higher-order pattern discovery, scalable network analysis on billion-edge graphs, and low-latency online inference using reusable, task-agnostic features derived from random walks, node-set sampling, and neighborhood hashing. In Thrust II, we extend the design principle to privacy-preserving relational learning, where structural dependencies in graphs often violate the gradient decoupling assumption in standard privacy learning mechanisms like differentially private stochastic gradient descent (DP-SGD). We propose the first differential private relational learning framework that disentangles sample dependencies through a tailored DP-SGD approach. This framework enables the private fine-tuning of large language models (LLMs) on sensitive graph data, effectively addressing associated computational complexities while achieving strong privacy-utility trade-offs. </p><p dir="ltr">By co-designing learning algorithms and system implementations, this dissertation demonstrates how graph-based AI can be both scalable and trustworthy, opening new avenues for learning from complex structured data in real-world applications.</p>

No description available.

Humans' incredible capacity to model the complexity of the physical world is possible because they cast this complexity as the composition of simpler entities and rules to process them. Extensive work in cognitive science indeed shows that human perception and reasoning ability is structured around objects. Motivated by this observation, a growing number of recent work focused on entity-centric approaches to learning representation and their potential to facilitate downstream tasks. In the first contribution, we show how an entity-centric approach to learning a transition model allows us to extract meaningful visual entities and to learn transition rules that achieve better compositional generalization. In the second contribution, we show how an entity-centric approach to generating graphs allows us to design a model for conditional graph generation that permits direct optimisation of the graph properties. We investigate the performance of our model in a prototype-based molecular graph generation task. In this task, called lead optimization in drug discovery, we wish to adjust a few physico-chemical properties of a molecule that has proven efficient in vitro in order to make a drug out of it. / L'incroyable capacité des humains à modéliser la complexité du monde physique est rendue possible par la décomposition qu'ils en font en un ensemble d'entités et de règles simples. De nombreux travaux en sciences cognitives montre que la perception humaine et sa capacité à raisonner est essentiellement centrée sur la notion d'objet. Motivés par cette observation, de récents travaux se sont intéressés aux différentes approches d'apprentissage de représentations centrées sur des entités et comment ces représentations peuvent être utilisées pour résoudre plus facilement des tâches sous-jacentes. Dans la première contribution on montre comment une architecture centrée sur la notion d'entité va permettre d'extraire des entités visuelles interpretables et d'apprendre un modèle du monde plus robuste aux différentes configurations d'objets. Dans la deuxième contribution on s’intéresse à un modèle de génération de graphes dont l'architecture est également centrée sur la notion d'entités et comment cette architecture rend plus facile l'apprentissage d'une génération conditionelle à certaines propriétés du graphe. On s’intéresse plus particulièrement aux applications en découverte de médicaments. Dans cette tâche, on souhaite optimiser certaines propriétés physico-chmiques du graphe d'une molécule qui a été efficace in-vitro et dont on veut faire un médicament.

No description available.

Les modèles de réseaux de neurones et en particulier les modèles profonds sont aujourd'hui l'un des modèles à l'état de l'art en apprentissage automatique et ses applications. Les réseaux de neurones profonds récents possèdent de nombreuses couches cachées ce qui augmente significativement le nombre total de paramètres. L'apprentissage de ce genre de modèles nécessite donc un grand nombre d'exemples étiquetés, qui ne sont pas toujours disponibles en pratique. Le sur-apprentissage est un des problèmes fondamentaux des réseaux de neurones, qui se produit lorsque le modèle apprend par coeur les données d'apprentissage, menant à des difficultés à généraliser sur de nouvelles données. Le problème du sur-apprentissage des réseaux de neurones est le thème principal abordé dans cette thèse. Dans la littérature, plusieurs solutions ont été proposées pour remédier à ce problème, tels que l'augmentation de données, l'arrêt prématuré de l'apprentissage ("early stopping"), ou encore des techniques plus spécifiques aux réseaux de neurones comme le "dropout" ou la "batch normalization". Dans cette thèse, nous abordons le sur-apprentissage des réseaux de neurones profonds sous l'angle de l'apprentissage de représentations, en considérant l'apprentissage avec peu de données. Pour aboutir à cet objectif, nous avons proposé trois différentes contributions. La première contribution, présentée dans le chapitre 2, concerne les problèmes à sorties structurées dans lesquels les variables de sortie sont à grande dimension et sont généralement liées par des relations structurelles. Notre proposition vise à exploiter ces relations structurelles en les apprenant de manière non-supervisée avec des autoencodeurs. Nous avons validé notre approche sur un problème de régression multiple appliquée à la détection de points d'intérêt dans des images de visages. Notre approche a montré une accélération de l'apprentissage des réseaux et une amélioration de leur généralisation. La deuxième contribution, présentée dans le chapitre 3, exploite la connaissance a priori sur les représentations à l'intérieur des couches cachées dans le cadre d'une tâche de classification. Cet à priori est basé sur la simple idée que les exemples d'une même classe doivent avoir la même représentation interne. Nous avons formalisé cet à priori sous la forme d'une pénalité que nous avons rajoutée à la fonction de perte. Des expérimentations empiriques sur la base MNIST et ses variantes ont montré des améliorations dans la généralisation des réseaux de neurones, particulièrement dans le cas où peu de données d'apprentissage sont utilisées. Notre troisième et dernière contribution, présentée dans le chapitre 4, montre l'intérêt du transfert d'apprentissage ("transfer learning") dans des applications dans lesquelles peu de données d'apprentissage sont disponibles. L'idée principale consiste à pré-apprendre les filtres d'un réseau à convolution sur une tâche source avec une grande base de données (ImageNet par exemple), pour les insérer par la suite dans un nouveau réseau sur la tâche cible. Dans le cadre d'une collaboration avec le centre de lutte contre le cancer "Henri Becquerel de Rouen", nous avons construit un système automatique basé sur ce type de transfert d'apprentissage pour une application médicale où l'on dispose d’un faible jeu de données étiquetées. Dans cette application, la tâche consiste à localiser la troisième vertèbre lombaire dans un examen de type scanner. L’utilisation du transfert d’apprentissage ainsi que de prétraitements et de post traitements adaptés a permis d’obtenir des bons résultats, autorisant la mise en oeuvre du modèle en routine clinique. / Neural network models and deep models are one of the leading and state of the art models in machine learning. They have been applied in many different domains. Most successful deep neural models are the ones with many layers which highly increases their number of parameters. Training such models requires a large number of training samples which is not always available. One of the fundamental issues in neural networks is overfitting which is the issue tackled in this thesis. Such problem often occurs when the training of large models is performed using few training samples. Many approaches have been proposed to prevent the network from overfitting and improve its generalization performance such as data augmentation, early stopping, parameters sharing, unsupervised learning, dropout, batch normalization, etc. In this thesis, we tackle the neural network overfitting issue from a representation learning perspective by considering the situation where few training samples are available which is the case of many real world applications. We propose three contributions. The first one presented in chapter 2 is dedicated to dealing with structured output problems to perform multivariate regression when the output variable y contains structural dependencies between its components. Our proposal aims mainly at exploiting these dependencies by learning them in an unsupervised way. Validated on a facial landmark detection problem, learning the structure of the output data has shown to improve the network generalization and speedup its training. The second contribution described in chapter 3 deals with the classification task where we propose to exploit prior knowledge about the internal representation of the hidden layers in neural networks. This prior is based on the idea that samples within the same class should have the same internal representation. We formulate this prior as a penalty that we add to the training cost to be minimized. Empirical experiments over MNIST and its variants showed an improvement of the network generalization when using only few training samples. Our last contribution presented in chapter 4 showed the interest of transfer learning in applications where only few samples are available. The idea consists in re-using the filters of pre-trained convolutional networks that have been trained on large datasets such as ImageNet. Such pre-trained filters are plugged into a new convolutional network with new dense layers. Then, the whole network is trained over a new task. In this contribution, we provide an automatic system based on such learning scheme with an application to medical domain. In this application, the task consists in localizing the third lumbar vertebra in a 3D CT scan. A pre-processing of the 3D CT scan to obtain a 2D representation and a post-processing to refine the decision are included in the proposed system. This work has been done in collaboration with the clinic "Rouen Henri Becquerel Center" who provided us with data

Ces dernières années, il y a eu un intérêt croissant dans le domaine de l'apprentissage profond pour le traitement du langage naturel. Plusieurs étapes importantes ont été franchies au cours de la dernière décennie dans divers problèmes, tels que les systèmes de questions-réponses, le résumé de texte, l'analyse des sentiments, etc. Le pré-entraînement des modèles de langage dans une manière auto-supervisé est une partie importante de ces réalisations. Cette thèse explore un ensemble de méthodes auto-supervisées pour apprendre des représentations de phrases à partir d'une grande quantité de données non étiquetées. Nous introduisons également un nouveau modèle de mémoire augmentée pour apprendre des représentations basées sur une structure d'arbre. Nous évaluons et analysons ces représentations sur différentes tâches. Dans le chapitre 1, nous introduisons les bases des réseaux neuronaux avant et des réseaux neuronaux récurrents. Le chapitre se poursuit avec la discussion de l'algorithme de rétropropagation pour former les réseaux neuronaux de flux avant, et la rétropropagation à travers l'algorithme de temps pour former les réseaux neuronaux récurrents. Nous discutons également de trois approches différentes dans le domaine de l’apprentissage de représentations, notamment l'apprentissage supervisé, l'apprentissage non supervisé et une approche relativement nouvelle appelée apprentissage auto-supervisé. Dans le chapitre 2, nous discutons des principes fondamentaux du traitement automatique du langage naturel profond. Plus précisément, nous couvrons les représentations de mots, les représentations de phrases et la modélisation du langage. Nous nous concentrons sur l'évaluation et l'état actuel de la littérature pour ces concepts. Nous finissons le chapitre en discutant le pré-entraînement à grande échelle et le transfert de l’apprentissage dans la langue. Dans le chapitre 3, nous étudions un ensemble de tâches auto-supervisées qui prend avantage de l’estimation contrastive bruitée afin d'apprendre des représentations de phrases à l'aide de données non étiquetées. Nous entraînons notre modèle sur un grand corpus et évaluons nos représentations de phrases apprises sur un ensemble de tâches du langage naturel en aval provenant du cadre SentEval. Notre modèle entraîné sur les tâches proposées surpasse les méthodes non-supervisées sur un sous-ensemble de tâches de SentEval. Dans les chapitres 4, nous introduisons un modèle de mémoire augmentée appelé Ordered Memory, qui présente plusieurs améliorations par rapport aux réseaux de neurones récurrents augmentés par pile traditionnels. Nous introduisons un nouveau mécanisme d'attention de Stick-breaking inspiré par les Ordered Neurons [shen et. al., 2019] pour écrire et effacer la mémoire. Une nouvelle cellule récursive à portes est également introduite pour composer des représentations de bas niveau en des représentations de haut niveau. Nous montrons que ce modèle fonctionne bien sur la tâche d'inférence logique et la tâche ListOps, et il montre également de fortes propriétés de généralisation dans ces tâches. Enfin, nous évaluons notre modèle sur les tâches (binaire et multi-classe) SST (Stanford Sentiment Treebank) et rapportons des résultats comparables à l’état de l’art sur ces tâches. / In chapter 1, we introduce the basics of feed forward neural networks and recurrent neural networks. The chapter continues with the discussion of the backpropagation algorithm to train feed forward neural networks, and the backpropagation through time algorithm to train recurrent neural networks. We also discuss three different approaches in learning representations, namely supervised learning, unsupervised learning, and a relatively new approach called self-supervised learning. In chapter 2, we talk about the fundamentals of deep natural language processing. Specifically, we cover word representations, sentence representations, and language modelling. We focus on the evaluation and current state of the literature for these concepts. We close the chapter by discussing large scale pre-training and transfer learning in language. In chapter 3, we investigate a set of self-supervised tasks that take advantage of noise contrastive estimation in order to learn sentence representations using unlabeled data. We train our model on a large corpora and evaluate our learned sentence representations on a set of downstream natural language tasks from the SentEval framework. Our model trained on the proposed tasks outperforms unsupervised methods on a subset of tasks from SentEval. In chapter 4, we introduce a memory augmented model called Ordered Memory with several improvements over traditional stack-augmented recurrent neural networks. We introduce a new Stick-breaking attention mechanism inspired by Ordered Neurons [Shen et.al., 2019] to write in and erase from the memory. A new Gated Recursive Cell is also introduced to compose low level representations into higher level ones. We show that this model performs well on the logical inference task and the ListOps task, and it also shows strong generalization properties in these tasks. Finally, we evaluate our model on the SST (Stanford Sentiment Treebank) tasks (binary and fine-grained) and report results that are comparable with state-of-the-art on these tasks.

<p>The availability of abundant multimodal data, including textual, visual, and sensor-based information, holds the potential to improve decision-making in diverse domains. Extracting data-driven decision-making information from heterogeneous and changing datasets in real-world data-centric applications requires achieving complementary functionalities of multimodal data integration, knowledge extraction and mining, situationally-aware data recommendation to different users, and uncertainty management in the open-world setting. To achieve a system that encompasses all of these functionalities, several challenges need to be effectively addressed: (1) How to represent and analyze heterogeneous source contents and application context for multimodal data recommendation? (2) How to predict and fulfill current and future needs as new information streams in without user intervention? (3) How to integrate disconnected data sources and learn relevant information to specific mission needs? (4) How to scale from processing petabytes of data to exabytes? (5) How to deal with uncertainties in open-world that stem from changes in data sources and user requirements?</p> <p><br></p> <p>This dissertation tackles these challenges by proposing novel frameworks, learning-based data integration and retrieval models, and algorithms to empower decision-makers to extract valuable insights from diverse multimodal data sources. The contributions of this dissertation can be summarized as follows: (1) We developed SKOD, a novel multimodal knowledge querying framework that overcomes the data representation, scalability, and data completeness issues while utilizing streaming brokers and RDBMS capabilities with entity-centric semantic features as an effective representation of content and context. Additionally, as part of the framework, a novel text attribute recognition model called HART was developed, which leveraged language models and syntactic properties of large unstructured texts. (2) In the SKOD framework, we incrementally proposed three different approaches for data integration of the disconnected sources from their semantic features to build a common knowledge base with the user information need: (i) EARS: A mediator approach using schema mapping of the semantic features and SQL joins was proposed to address scalability challenges in data integration; (ii) FemmIR: A data integration approach for more susceptible and flexible applications, that utilizes neural network-based graph matching techniques to learn coordinated graph representations of the data. It introduces a novel graph creation approach from the features and a novel similarity metric among data sources; (iii) WeSJem: This approach allows zero-shot similarity matching and data discovery by using contrastive learning<br> to embed data samples and query examples in a high-dimensional space using features as a novel source of supervision instead of relevance labels. (3) Finally, to manage uncertainties in multimodal data management for open-world environments, we characterized novelties in multimodal information retrieval based on data drift. Moreover, we proposed a novelty detection and adaptation technique as an augmentation to WeSJem.<br> </p> <p>The effectiveness of the proposed frameworks, models, and algorithms was demonstrated<br> through real-world system prototypes that solved open problems requiring large-scale human<br> endeavors and computational resources. Specifically, these prototypes assisted law enforcement officers in automating investigations and finding missing persons.<br> </p>