Differences

This shows you the differences between two versions of the page.

--- teaching:mfe:ia [2011/03/22 14:48]
stuetzle
+++ teaching:mfe:ia [2016/03/11 11:46]
stuetzle [Optimising Ant Colony Algorithms for Performance]
@@ Line 1: / Line 1: @@
-====== MFE 2011-2012 : Intelligence Artificielle ======
+====== MFE 2015-2016 : Intelligence Artificielle ======
 ===== Introduction =====
@@ Line 7: / Line 7: @@
 Ces sujets sont prêt à être encadrer, mais il va s'en dire qu'ils ne sont pas uniques. Les étudiants sont vivement encouragés à prendre contact avec Hugues Bersini (bersini AT ulb.ac.be) ou Marco Dorigo (mdorigo AT ulb.ac.be) afin de discuter de l'une ou l'autre initiative inspirée pouvant faire l'objet dun autre sujet de MFE ou de préciser le cadres, le contenu et les attentes relatives au sujets présentés.
-===== Exploration, exploitation and evaluation of the .Net Linq library for the problem of object permanence. =====
-The MFE will consist in a deep exploration of the
+===== Développer un programme informatique permettant une analyse statistique en vue de  l'évaluation d'un module psychothérapeutique. =====
-very new microsoft Linq library which aims at resolving
-the classical problem of mapping between the OO and the relational
-and the XML world. The different additions of .Net necessary
-to the implementation of this library will be studied. The
-library will also be tested in terms of robustness and performance
-as compared with the previous existing solutions coming from other technological platforms (Java, PHP, ...). This memoire will be a follow up of a previous memoire.
+Ce mémoire se fera en collaboration avec l'équipe médicale du centre pour l'anorexie et la boulimie de l'hôpital Erasme. Il consistera en l'analyse informatisée des données récoltées lors d'entretiens avec le patient et sa famille au cours du traitement. Les données sont actuellement stockées dans dans une base de données SPSS.  Le mémoire consistera pour l'essentiel au traitement de ces données par des approches "Machine Learning" et "Data Mining" dans une perspective de Quality Management.
   * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
@@ Line 22: / Line 16: @@
 ===== Développer un programme informatique permettant une analyse statistique en vue de  l'évaluation d'un module psychothérapeutique. =====
-Ce mémoire se fera en collaboration avec l'équipe médicale du centre pour l'anorexie et la boulimie de l'hôpital Erasme. Il consistera en l'analyse informatisée des données récoltées lors d'entretiens avec le patient et sa famille au cours du traitement.
+Ce mémoire se fera en collaboration avec l'équipe médicale du centre pour l'anorexie et la boulimie de l'hôpital Erasme. Il consistera en l'analyse informatisée des données récoltées lors d'entretiens avec le patient et sa famille au cours du traitement. Les données sont actuellement stockées dans dans une base de données SPSS.  Le mémoire consistera pour l'essentiel au traitement de ces données par des approches "Machine Learning" et "Data Mining" dans une perspective de Quality Management.
   * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
+===== Contribution au développement de la plateforme Big Data d'IRIDIA =====
+Dans de nombreux projets d'enseignement et d'industrie dans lesquels IRIDIA se trouvent impliqués, des demandes pressantes nous sont faites de formation et d'utilisation des outils Big Data (Map-Reduce, Cloudera, Hue, Hive, Pig, Elastic Search, HBase). Nous répondons tant bien que mal à ces demandes et tentons d'équiper notre cluster avec les logiciels demandés. Le mémoire consisterait en une étude comparative de l'existant en terme surtout logiciel et la mise en place d'un cours mettant en évidence les avantages de l'une ou l'autre solution en ce compris par rapport à des solutions plus conventionnels de type BD relationnelle.
+  * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
+===== Contribution à l'amélioration de la plateforme génomique In Silico DB =====
+Une nouvelle spin-off a vu le jour depuis un an à IRIDIA: In Silico DB (https://insilicodb.org/) mettant à disposition sous une forme aisément exploitable des centaines de milliers d'échantillons de données génomiques permettant un meilleur diagnostic des maladies d'origine génétique et une meilleure compréhension de la biologie moléculaire. L'équipe qui s'en occupe a un besoin pressant de développeurs informatiques permettant d'en améliorer l'interface. Des connaissances en programmation Web sont souhaitées. La migration du système de stockage sur un serveur Big Data est aussi un des objectifs actuellement poursuivis par l'équipe en place et permettra au mémorant de se familiariser avec ces nouvelles technologies.
+  * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
+===== Etude pratique et expérimentale du langage de programmation F# =====
+Depuis quelques années, Microsoft met en avant un nouveau langage de programmation F# créé dans le sillage des langages dits déclaratifs ou fonctionnels. Il semblerait que ce langage soit idéal pour le traitement des données. Le MFE consistera en une étude expérimentale de ce langage et un comparatif avec les langages de programmation aujourd'hui les plus usités.
+ * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
+===== Etude de l'algorithme du Deep Learning =====
+Les réseaux de neurones multicouches sont redevenus très à la mode depuis que Google les utilise massivement pour le traitement automatique d'images et de vidéos. Nous avons à IRIDIA étudié et réalisé plusieurs algorithmes d'apprentissage de ces réseaux multicouches. Le mémoire consistera en une comparaison des algorithmes de deep learning tels ceux utilisés chez Google et les alternatives que nous proposons ici à IRIDIA depuis plusieurs années.
+ * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
 ===== Etude et réalisation orientée objet d'une cellule minimale =====
@@ Line 34: / Line 51: @@
 élémentaire et sa membrane. L'idée est de réaliser le logiciel minimal capable de simuler un organisme vivant. Cette cellule devra être capable
 de croître et de spontanément se dupliquer. Il fera suite
-à une succession de MFE déjà réalisés ces dernières années.
+à une succession de MFE déjà réalisés ces dernières années. Parmi les amélirations possibles sont en cours la parallélisation du programme suivant une technologie Big Data ou le recours aux processeurs GPU.
   * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
+===== Réorganisation sous forme OO et UML d’un code de simulation climatique =====
+Ce mémoire se déroulera en collaboration avec le professeur Jean-Pascal van Ypersele de l’UCL, vice président du GIEC, groupe de recherche sur l’évolution climatique. La plupart des codes de simulation climatique sont rédigés en Fortran en exploitant peu les principes de la programmation OO. Ce mémoire consistera en la sélection d’un logiciel de simulation climatique assez simple, plutôt à vocation didactique, et sa réécriture sous forme OO, en faisant un recours intensif aux diagrammes UML et aux Design Patterns.
+  * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
+===== Réorganisation sous forme OO et UML d’un code de contagion systémique d'un réseau de crédits interbancaire  =====
+La crise financière actuelle a permis de mettre en lumière les risques de contagion systémique liés à la faillite de certaines banques. En effet, la plupart du temps, les banques forment entre elles un réseau de crédit interbancaire qui, à la fois les rend plus solides, mais aussi plus vulnérables à la défection de l’une ou l’autre. De nombreux logiciels ont été écrits afin d’étudier plus en détail ce risque. Le mémoire consistera en la sélection d’un de ces logiciels déjà clairement identifiés et sa réécriture sous forme OO, en faisant un recours intensif aux diagrammes UML et aux Design Patterns.
+  * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
 ===== Data/text mining - Traitement automatique de documents sur base de leur contenu =====
@@ Line 66: / Line 94: @@
   * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
+===== Etude de la topologie de réseaux d'acteurs extraits à partir de romans célèbres  =====
+De plus en plus de scientifiques sont convaincus qu’une même topologie de réseaux (c'est-à-dire la manière dont les nœuds en sont connectés) se retrouve dans de nombreux réseaux, pourtant extraits de réalités très diverses (Web, Internet, réseaux sociaux, biologiques, épidémiques). Cette topologie leur conférerait des propriétés intéressantes comme une plus grande robustesse ou une communication réduite entre les nœuds. Il est possible de construire des réseaux lexicaux à partir d’un document quelconque, par exemple, en connectant deux mots qui apparaissent dans une même phrase. Le MFE consistera en un développement logiciel ayant pour but la réalisation automatique d'un réseau de personnages de romans (Harry Potter, les Misérables et autres) à partir des dialogues présents dans ces romans. Les liens seront également pondérés comme résultat d'une analyse de sentiments faite à partir de ces mêmes dialogues. On procédera ensuite à l’étude automatisée de leur topologie : distance inter-nœuds, degré de clustering, etc …
+  * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
 ===== Expérimentation des designs patterns pour la modélisation de systèmes biologiques complexes =====
-Tout bon informaticien se doit aujourd’hui de maîtriser ces recettes de conception OO que sont les designs patterns. Au-delà des langages de programmation ou de modélisation (UML), ils sont devenus le sujet d’étude et de développement le plus prisé de la communauté informatique. Leur maîtrise permet à ces mêmes informaticiens d’attaquer la simulation de procédés complexes avec plus de facilité. Le MFE consistera en la mise en pratique de ces designs patterns pour la modélisation de systèmes biologiques complexes comme le système immunitaire ou les mécanismes de régulations génétiques. Le travail devrait déboucher sur une adaptation de ces mêmes designs patterns au monde et aux problèmes de la biologie.
+Tout bon informaticien se doit aujourd’hui de maîtriser ces recettes de conception OO que sont les designs patterns. Au-delà des langages de programmation ou de modélisation (UML), ils sont devenus le sujet d’étude et de développement le plus prisé de la communauté informatique. Leur maîtrise permet à ces mêmes informaticiens d’attaquer la simulation de procédés complexes avec plus de facilité. Le MFE consistera en la mise en pratique de ces designs patterns pour la modélisation de systèmes biologiques complexes comme le système immunitaire ou les mécanismes de régulations génétiques. Le travail devrait déboucher sur une adaptation de ces mêmes designs patterns au monde et aux problèmes de la biologie. Ce mémoire se fera en collaboration avec l'institut de recherche de Microsoft à Cambridge qui a déjà supervisé les mémorants précédents.
   * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
@@ Line 87: / Line 120: @@
   * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
@@ Line 94: / Line 128: @@
   * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
+===== Mise en place de solutions Big Data et Elastic Search pour les réseaux bibliométriques  =====
+Les publications scientifiques doivent se citer entres elles. Elles constituent donc un immense réseau de citations. Nous étudions à IRIDIA la nature de ce réseau et l'impact que peut avoit une publication scientifique en suivant dans le temps le nombre de nouvelles publications qui citent cette dernière. Comme ces réseaux sont de taille gigantesque (des millions de publications), des solutions technologiques harwarde et software de type Big Data et Elastic Search sont mises en place et expérimentées ici à IRIDIA. Le mémoire consistera en un développement et une expérimentation des ces possibles solutions.
+* Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
 ===== Mise au point d’un langage de modélisation de systèmes biologiques inspiré des diagrammes de classe et d'état/transition UML =====
@@ Line 99: / Line 141: @@
   * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
 ===== Comparaison via la simulation informatique d'une économie de marché de nature concurrentielle et une autre plus redistributive=====
 Les économistes nous assènent à l'envi que l'économie se doit d'être compétitive et parfaitement concurrentielle. Est-ce si vrai ? L'économie de marché ne peut-elle exister que sur un mode concurrentiel pour assurer au mieux le bonheur du plus grand nombre d'agents économiques ? Nous adresserons cette question par l'entremise de modèles économiques multi-agents mettant en présence des producteurs,acheteurs, consommateurs et vendeurs, et les faisant se comporter d'abord sur un monde compétitif (économie de marché de type enchère) et ensuite aléatoire. Nous étudierons   la manière dont le bien-être cumulé par les agents consommateur est distribué parmi eux. Ce mémoire fait suite à un mémoire réalisé par un étudiant de Solvay l'année passée et donc il pourra repartir d'un logiciel existant.
+  * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
+===== Etude des instabilités dynamiques des marchés boursiers =====
+Malgré la théorie économique voyant dans le marché un processus auto-régulé et stable, le fonctionnement de la bourse et de la finance se caractérisent par d'incessantes instabilités dynamiques: bulles spéculatives et autres... Ce MFE aura pour objet une modélisation d'un marché boursier très simplifié dans lesquels seront pris en compte les mimétismes des "traders" souvent responsables de phénomènes de feedbacks positifs menant à ces instabilités.
   * Contact : [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
@@ Line 127: / Line 176: @@
   * Contacts :
     * [[http://iridia.ulb.ac.be/~mdorigo|Marco Dorigo (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stuetzle (IRIDIA)]]
+    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stützle (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~manuel|Manuel López-Ibáñez (IRIDIA)]]
-===== Optimising Ant Colony Algorithms for Performance ======
-Ants have inspired a number of computational techniques and among the most successful is ant colony optimization (ACO). ACO is an optimization technique that can be applied to tackle a wide variety of computational problems that arise in computer science, telecommunications, and engineering.
-The goal of this project is to improve the performance of ACO algorithms by investigating and testing various implementation techniques: intrinsic functions (MMX/SSE floating-point operations), CPU cache effects, or GPU programming.
+===== Design of a graphical interface for an automatic configuration tool. =====
-Required skills: knowledge of C programming. Some knowledge about computer architecture.
+Optimization algorithms have a number of parameters that strongly
+affect their efficiency. For many years the setting of these
+parameters was done by hand; a tedious task that requires a lot of
+human involvement. Nowadays, some tools are available to automatize this task by considering the setting of the parameters as a "meta"-optimization problem. One of these tools for automatic configuration (the irace package: http://iridia.ulb.ac.be/irace) has been developed at IRIDIA, and has been already applied successfully to many algorithms. The goal of this project is to design a graphical interface on top of the existing software, to help the user to set-up his particular tuning problem, to visualize information about the tuning process while it is on-going and when it has completed, and to integrate statistical tools for the analysis of the tuner results.
+The student will have to implement a Graphical front-end on top of the
+existing software implemented in R, using a cross-platform library
+such as Qt (http://qtinterfaces.r-forge.r-project.org/).  Some
+additions to the original software may be required, and the student
+will have to work in collaboration with the team of developers of
+irace at IRIDIA.
   * Contacts :
-    * [[http://iridia.ulb.ac.be/~mdorigo|Marco Dorigo (IRIDIA)]]
+    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stützle (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stuetzle (IRIDIA)]]
     * [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~manuel|Manuel López-Ibáñez (IRIDIA)]]
+    * [[http://iridia.ulb.ac.be/~lperez|Leslie Perez (IRIDIA)]]
-===== Graphical Tools for analysing Multi-Objective Data. =====
-In multi-objective problems, not only one objective function must be minimised but several, often conflicting, objectives must be taken into account. The result is often a set of solutions modelling the trade-off between the objectives of the problem.
-The goal of this project is to develop graphical tools to interactively examine and compare the results of algorithms for multi-objective problems.
+===== Feature Extraction and Automatic Algorithm Selection. ======
-Required skills: The candidate should be well acquainted with programming in object oriented languages.
+The performance of (Stochastic Local Search) algorithms for a given problem depends on the algorithm design and on the setting of the algorithm's parameter. Given a heterogeneous set of instances for a given problem a good algorithm design (or parameter configuration) for one instance is not necessary the best design for all instances. On the contrary a tuning of an algorithm on a specific family of similar instances may affect negatively its performance on other families of instances.
+The thesis will focus on devising automatic methods for extracting features from the instances, select the relevant features, and learning (in the framework of multi-class classification) the
+relationship, if there is one, between the instances features and the best algorithm for the instance. The results will be instrumental for algorithm selection or the creation of portfolios of complementary algorithms suitable for large sets of diverse instances for a given problem.
+Required skills: good knowledge of C or C++ programming and of a scripting language (e.g., python); good knowledge of machine learning methods would also be helpful.
   * Contacts :
-    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stuetzle (IRIDIA)]]
+    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stützle (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~manuel|Manuel López-Ibáñez (IRIDIA)]]
+    * [[http://code.ulb.ac.be/iridia.people.php?id=1393|Alberto Franzin (IRIDIA)]]
-===== Stochastic local search algorithms for weighted maximum clique problems. ======
-The Maximum Clique Problem is an NP-hard combinatorial optimisation problem that asks to find the biggest completely
+===== Formal verification of a swarm robotics behavior through statistical model checking =====
-connected component of a graph. It has relevant applications in information retrieval, computer vision, social network
-analysis, computational biochemistry, bioinformatics and genomics.
-Among the possible generalisations of the problem there is the Vertex Weighted and Edge Weighted Maximum Clique which asks to find the clique of maximum weight. Being generalisations they are also NP-hard. The goal of the project is to devise heuristic algorithms or adapt existing algorithms of the Maximum Clique for weighted version.
+The goal of this thesis is to apply statistical model
+checking to formally verify properties of a collective behavior of a
+robot swarm. Verifying that a system behaves as desired in all
+possible situations is necessary when autonomous robots are involved.
+This is particularly true in swarm robotics systems, where the
+interactions of large number of individuals can result in behaviors
+difficult to predict. Model checking is a common technique to formally
+prove properties of a system. However, its results are limited to
+small systems, because medium-sized or large systems are
+computationally impossible to analyze.
-Required skills: good knowledge of C or C++ programming.
+This thesis is will explore the application of a novel model checking technique, called statistical model checking, to formally verify a swarm robotics system. A collective behavior will be firstly implemented in
+simulation and then analyzed through statistical model checking.
+Required skills: the candidates should be acquainted with C/C++
+programming and have a working knowledge of the English language.
-  * Contacts :
+  * Contacts : [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]] and Manuele Brambilla (IRIDIA)
-    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stuetzle (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~fmascia|Franco Mascia (IRIDIA)]]
+===== ROBO-POINTER -- A New Device for Human-Robot Interaction =====
-===== Stochastic Local Search heuristics for solving NP-complete puzzles. ======
+In this project, the student will design and implement a novel human-robots interaction system.
+The human uses a special laser pointer (called robo-pointer), and by simply pointing to a specific robot he can select it, check its status on a dedicated monitor and send to the robot instructions.
-This project is about single player games (puzzles) and the design of algorithms for tackling hard combinatorial optimisation problems.
+Now, our arena is equipped with a multi-camera tracking system that allows to detect the 3D position of the robots.
-Example puzzles are: <a href="http://en.wikipedia.org/wiki/Light_Up">Light Up</a>, <a href="http://en.wikipedia.org/wiki/Mastermind_(board_game)">Mastermind</a>, <a href="http://en.wikipedia.org/wiki/Minesweeper_(video_game)">Minesweeper</a>, etc.
+The student will design a radio sensor network to detect the 3D position of the robo-pointer.
+The two pieces of information will be combined to trigger the selection of the pointed robot.
-The student will learn how to design and implement a Stochastic Local Search algorithm to solve NP-complete puzzles. The student will also learn how to analyse the performaces of the algorithm and perform statistically sound comparisons with the other algorithms available in literature.
+* Contacts : [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]] and Andreagiovanni Reina (IRIDIA)
-Required skills: good knowledge of C or C++ programming.
+===== Let a robot swarm get around home! =====
-  * Contacts :
+Robot swarms have very limited knowledge of their environment and therefore, moving around a complex environment (i.e., room with obstacles, building, ...) is a complex task. The idea of this project is to allow an overhead multi-camera tracking system -- that has a complete vision and knowledge of the environment -- to interact with a robot swarm in order to help the swarm move to a specific goal in the environment (for instance, moving from the kitchen to a bedroom of a house).
-    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stuetzle (IRIDIA)]]
+The multi-camera tracking system's software is already developed. The student will be supposed to (i) develop a GUI that allows a human operator to specify a goal in the environment, (ii) design and implement a communication protocol (between the tracking system and the robot swarm) that guarantees efficiency and scalability and finally, (iii) test the interaction system in a simulated environment.
-    * [[http://iridia.ulb.ac.be/~fmascia|Franco Mascia (IRIDIA)]]
+* Contacts : [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]], Gaëtan Podevijn, and Andreagiovanni Reina (IRIDIA)
-/*
-===== Swarm robotics using the e-puck platform =====
-The e-Puck is a robot developed by the Ecole Polytechnique Fédérale de Lausanne, Switzerland. It is equipped with a dsPIC micro-controller, it has an RS232 and a bluetooth interface, 8 infrared proximity sensors, a 3 axis accelerometer, 3 microphones and a speaker, a color camera with a resolution of 640x480 pixels and 8 red leds for displaying patterns.
+===== Collective Decision Making with Heterogeneous Agents =====
-In the last years, a number of projects carried out at IRIDIA developed a set of tools and a fully functional platform to work efficiently with  e-puck robots. In particular, a precise description of the properties of the robots, software libraries and an accurate simulator are now available.  A number of controllers were developed and successfully tested on the robots.
+Swarm robotics is an interesting approach to the coordination of hundreds of robots as it promotes the realization of systems which are scalable, robust and flexible.
+The master thesis will study how to provide a swarm system with the cognitive capability of collective decision making.
+Each agent has partial knowledge of the available alternatives and of their quality estimate, however the swarm, as a whole, is able to decide for the best option. Recently, numerous works have studied strategies and algorithms to implement this process in distributed systems (often taking inspiration from biology, e.g., bees or cockroaches behaviour). One of the common characteristic of these works is that all the agents of the swarm has the same behaviour. In the Master Thesis project, the student will study how heterogeneity influences the global outcome. We will consider heterogeneity both in the individual behaviour (for instance, robots can estimate diﬀerent option characteristics) and in the interaction network.
-The goal of the project ``Swarm robotics using the e-puck platform'' is to design and carry out experiments of swarm robotics that are typically bio-inspired and involve several robots.  Possible experiments include p2p communication networks for path finding, flocking for exploration, transport of objects and aggregation of robots. 12 e-Puck will be available for the project.
+In practice, the student is supposed to (i) model the collective decision making problem
+(ii) design and implement multi-agents simulations, and (iii) analyse and discuss the obtained
+results. Depending on the student skills and preferences, the work can focus more on theoretical
+aspects, thus favouring the modelling and analysis of the problem, or can be more practical,
+thus centring the thesis on the multi-agent implementation part. Possibly, a more practical
+thesis could result (depending on the student skills) in the implementation of a real world
+demonstrator with a swarm of up to 100 robots.
-The project is tightly connected to the research in swarm robotics carried out at IRIDIA and in particular to the EU funded //Swarmanoid// project, the aim of which is to study new approaches to the design and implementation of self-organizing and self-assembling artifacts. See [[http://www.swarmanoid.org]] for more details.
+* Contacts : [[http://iridia.ulb.ac.be/~mdorigo|Marco Dorigo]] and Andreagiovanni Reina (IRIDIA)
-Required skills: The candidates should be acquainted with C/C++ programming and have a working knowledge of the English language.
-  * Contacts :
+===== Online Framework to Compose Robotic Solutions =====
-    * [[http://iridia.ulb.ac.be/~mdorigo|Marco Dorigo (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari (IRIDIA)]]
-*/
-===== Self-organized task allocation in swarm robotics =====
+The design of swarm robotics systems is a challenging task due to difficulties to predict and model swarm systems.
+The most common approach is a trial-and-error process where the developer tries solutions, tests them in simulation and corrects the code to eventually achieve the desired global outcome.
-Swarm robotics is an innovative branch of collective robotics that aims at designing robot behaviors by taking inspiration from social animals, such as ants and bees. "Task allocation" in such robotic swarms is the problem of "who is doing what job and when?" Obviously, this problem of assigning jobs to a whole swarm robots can be very difficult, especially when using many robots that cannot communicate with each other on a global level.
+The project aims to ease this design process: the developer have access to the code of already solved problems (called code-modules), and he can graphically combine the code-modules through a graphical interface to create his own solution.
-The goal of the project is to implement new algorithms for solving this problem on the e-Puck robot and run extensive experiments with this robot in various environments. The project will involve experimentation with about 30 real e-Pucks. The project is tightly connected to the research in swarm robotics carried out at IRIDIA.
+The student is supposed to design this novel online framework that offers two main functionalities: (1) to compose new solutions combining the existing code-models to solve new open problems, and (2) to code new code-modules and submit them to the online system.
-Required skills: The candidates should be acquainted with C/C++ programming and have a working knowledge of the English language.
+* Contacts : [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]], Andreagiovanni Reina, and Gianpiero Francesca (IRIDIA)
-  * Contacts: [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]], Marco Dorigo, Arne Brutschy, Giovanni Pini (IRIDIA)
-===== Studying collaboration between flying robots and ground-based robots =====
+===== UML for Swarm robotics: formal specification of a collective behavior =====
-In previous studies, it has been shown that multiple ground-based robots can autonomously form various patterns by attaching to each other. These robots used simple rule sets and local communication to form pre-defined or random patterns. In this thesis, the student will study how flying robots can collaborate with ground-based robots to select and control the pattern formation process. The student will implement the results of his study and various other algorithms that would facilitate such a collaboration. In order to gain a sound understanding of the matter, the student will first study and benchmark collaboration techniques used in existing robotic systems including flying and ground-based robots.
+Swarm robotics is an interesting approach to the
+coordination of hundreds of robots as it promotes the realization of
+systems which are scalable, robust and flexible. However, up to now,
+swarm robotics application has been quite limited, also due to the
+lack of an engineering approach to its development.
+In particular, formal specification has not been applied yet to swarm
+robotics systems.
-A possible candidate student must be very motivated, ready to invest extra hours into the thesis, and have a good grasp of C++.  The working  language is English.
+In this thesis, we will explore possible ways to formally specify
+swarm robotics systems. As a starting point we will consider UML and
+UML extensions like AUML and UML for multi-agent systems. If
+necessary, we will develop a specific extension for swarm robotics
+systems. Once the preliminary work is done we will consider an
+example, perform formal specification of a task and then implement the
+system in simulation.
-  * Contacts: [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]], Marco Dorigo, Nithin Mathews (IRIDIA)
+Required skills: the candidates should be acquainted with C/C++
+programming, have a good knowledge of formal specification and UML,
+and have a working knowledge of the English language.
-===== Adaptive collective alignment with a swarm of e-puck robots =====
+  * Contacts : [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]] and Manuele Brambilla (IRIDIA)
-Flocking is a fascinating behavior that birds are able to achieve without a leader or a common frame of reference. Moreover, in some cases, the group goes in the correct direction even if only a small proportion of the group knows the goal direction. This allows birds to avoid a predator even if only a subset of the flock sees it. We want to study one of the most interesting aspects of this mechanism, that is how a group can align collectively to a common direction and change this direction over time according to some stimuli perceived only by a small minority of individuals.
-The goal of this project is to apply a methodology, so far studied only in simulation, to the e-puck robots, in order to tackle the adaptive collective alignment problem. A group of e-pucks has to reach consensus and turn to a common random heading direction, using a common light source as reference point. Furthermore, when an obstacle is perceived by a small minority of the group, consensus should be achieved in order to align to a new direction which allows them to avoid the obstacle.
+===== A GUI for debugging the behavior of a robot swarm =====
-Required skills: The candidates should be acquainted with C/C++ programming and have a working knowledge of the English language.
+Debugging a robot swarm is a complex and difficult task.
+The desired behavior of the swarm is the result of the complex
+non-linear interactions of tens or hundreds of robots. When
+implementing a swarm robotics system, very often it is necessary to
+analyze individually the output of the execution of each robot, a very
+long and boring process. Since the goal of the developer is to obtain
+a specific collective behavior, it would be better to debug the system
+at the collective level and, only if necessary, at the individual
+level.
+In this thesis we will analyze a possible way to debug the collective
+behavior of a swarm of robots, using macroscopic and microscopic
+modeling. The goal is to develop a GUI that shows the state of the
+collective behavior of the system, and if the user requires it, the
+state of a single robot. We will start with a version of the debugging
+GUI that interface with the ARGoS simulator and eventually one that
+interfaces with the real robots.
-  * Contact: [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]], Marco Dorigo, Eliseo Ferrante, Ali Emre Turgut (IRIDIA)
+Required skills: the candidates should be acquainted with C/C++
+programming, GUI programming (QT/C++ or QT/Python or Java) and have a
+working knowledge of the English language.
-===== Scalable aggregation in swarm robotics without global information or environmental clues =====
-Several studies in biology have shown that group of social insects are able to gather to a particular spot. This process is usually driven by environmental clues such as shadows projected by a shelter (cockroaches) or temperature gradients (bees). These studies have been a source of inspiration for several algorithms in swarm robotics. Is it possible to achieve the same result without an environmental clue? Do we need global information in order to let a group of robot gather in one place?
+  * Contacts : [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]] and Manuele Brambilla (IRIDIA)
-The goal of this project is to study how to solve an aggregation task without relying on environmental clues or global signaling. The problem can be seen as an exploration-exploitation trade-off tackled by a single robot. The robot has to select between keeping exploring, that is, finding the the largest aggregate, or exploiting, that is join a previously created aggregate. The study will be conducted only in simulation and will concern comparing different approaches for decision making or different communication strategies.
-Required skills: The candidates should be acquainted with C++ programming and have a working knowledge of the English language.
-  * Contact: [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]], Marco Dorigo, Eliseo Ferrante, Ali Emre Turgut (IRIDIA)
-===== A comparison of decision-making strategies for adaptive foraging in swarm robotics =====
+===== A virtual machine for mobile code in a swarm of robots =====
-Group of social insects are able to efficiently find the (shortest) path to the a food source and even to differentiate between the quality of two food sources. Studies with ants showed that this mechanism is driven by the perception of stimuli from chemical substances like pheromone. Moreover ants are able to collectively modify their choices if there are changes in the environment, that is, if a source becomes better than another. These ideas have been a source of inspiration for several algorithms in swarm robotics which solves a similar problem (retrieval of objects) by using different types of stimuli such as the encounter rate of objects.
+Mobile code is a technology whereby nodes in a network of
+computing nodes exchange code. In other words, code migrates from
+machine to machine like an agent navigating an environment. Mobile
+code is a promising technology for swarm robotics because it would
+enable a new, novel type of robot-to-robot interaction. The aim of this
+project is produce a simple, yet high-performance virtual machine to
+support code exchange in a swarm of robots. A simple experiment with
+the robots demonstrating the capabilities of the VM will be performed.
-The goal of this project is to perform a study on how to solve a foraging task in which robots have to choose between staying at the nest or go foraging for different energy sources. The optimal strategy might change over time. What happens if all the robots go to the best source? Will these "traffic jams" slow the process? Is it possible to avoid this problem? What if source quality changes over time? The study will be conducted only in simulation and will concern comparing different approaches and different metrics to measure stimuli.
+Required Skills: Good knowledge of C
-Required skills: The candidates should be acquainted with C++ programming and have a working knowledge of the English language.
-  * Contact: [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]], Marco Dorigo, Eliseo Ferrante, Manuele Brambilla (IRIDIA)
+  * Contact: [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]] and Carlo Pinciroli (IRIDIA)
-===== Kaleidoscope: Creating temporal motion patterns in a swarm of robots =====
-In swarm robotics, agents are programmed in such a way that local actions and simple interactions among agents result in complex, swarm-level dynamics. At present, the design of swarm robotic control systems is more of a craft than a science, mainly because significant design patterns are still to be identified and studied. This project aims to discover and study temporal patterns in robot motion, and subsequently to encode them into reusable design patterns. Each robot is assumed to possess a limited set of capabilities, such as the ability to change body color and to perceive other robots and their
-colors in a short range. Individual controllers are derived from a very simple but powerful mathematical model. The work of the student will be to code and analyze robot controllers, both with simulated and real robots. The most important required skills are a good knowledge of C and C++ and no fear of mathematics. The working language is English.
+===== Swarmscope =====
-  * Contact: [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]], Marco Dorigo, Carlo Pinciroli (IRIDIA)
+One the main problems in the development of swarm robotics
+systems is the difficulty of producing, analyzing and debugging code for
+large distributed systems. The aim of this project is to produce a set of
+innovative tools to aid the development of complex swarm robotics
+systems. The produced tools will involve new, creative visualization
+methods and media, novel human-robot swarm interaction and effective
+debugging tools.
+Required Skills: Good knowledge of C++ and Qt4
+  * Contact: [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]] and Carlo Pinciroli (IRIDIA)
+===== Automatic fitness function definition in evolutionary robotics =====
+Evolutionary robotics is a fascinating approach to the design of robot controllers that takes inspiration from natural evolution.
+In order to obtain a robot that is able to perform a desired task, the evolutionary robotics approach considers a population of robots that evolves in time. Each robot is characterized by a genotype that defines somehow its behavior. Each robot is evaluated according to a fitness function that measures the ability of the robot to perform the desired task. Robots with a low fitness are eliminated. Robots with a high fitness remain in the population and generate offsprings -- e.g., robots with a similar genotype obtained via mutation and/or cross-over. Through this process, generation by generation, the evolutionary robotics approach is able to obtain robots that present higher and higher fitness and that are therefore able to perform the desired task more and more effectively.
+One of the main open problems in evolutionary robotics is that the definition of an appropriate fitness function is a very complex, labor-intensive, and time-consuming activity that requires the attention of an expert researcher.
+The goal of this master thesis is to devise an automatic method to define a fitness function in order to obtain a robot that is able to perform a desired task. This automatic method will be based on machine learning and metaheuristic algorithms. In particular, it will draw ideas from the fields of reinforcement learning and of on-line adaptation of parameters in optimization algorithms.
+Required skills: The candidates should be acquainted with C/C++ programming and have a working knowledge of the English language.
+* Contact: [[http://iridia.ulb.ac.be/~mbiro|Mauro Birattari]], Marco Dorigo, Vito Trianni (IRIDIA)
-===== Simulation et optimisation de trafic routier =====
+===== Evolution of Modular Controllers for Simulated and Real Robots =====
-Mentis, spin-off du laboratoire IRIDIA, est une société de consultance
+The goal of this master thesis is investigating how modularity in a robot controller can influence the quality of the behaviours obtained through artificial evolution.
-spécialisée en Data Mining et en Text Mining.  Mentis cherche
+Similarly to the nervous system that can be divided in central and peripheral, the project will study a modular architecture for neural network controllers. The peripheral modules encode the information coming from the sensory subsytems or going to the motor apparatus. The central system encodes the behavioural rules that map sensations to actions. The project will study methods to develop the peripheral modules by maximising the information transfer from the sensory input and to the motor output, on the basis of measures derived from Information Theory.
-actuellement des mémorants pour lancer diverses études d’optimisation
+The project will involve experimental activities with both simulated and real robots, and will investigate both individual and collective behaviours.
-dans le domaine de la simulation de trafic routier.
-Dans le cadre d’un projet pour un de ses clients, Mentis travaille
+Required skills: The candidates should be acquainted with C/C++ programming and have a working knowledge of the English language.
-actuellement dans le domaine de la simulation de trafic routier.
-L’objectif du mémoire proposé consiste à lancer plusieurs études
-d’optimisation pour évaluer l’impact de différentes politiques
-routières sur le trafic.  Il sera demandé à l’étudiant de mettre en
-œuvre diverses techniques d’optimisation afin de déterminer les
-politiques routières optimales. Une grande partie du mémoire sera
-faite dans les bureaux de Mentis ainsi qu’en interaction directe avec
-le client.
+* Contact: [[http://iridia.ulb.ac.be/~vtrianni|Vito Trianni]], Marco Dorigo (IRIDIA)
-  * Contacts:
-    * [[psener@mentis-consulting.be|Pierre Sener (Mentis)]]
+===== Localisation and tracking of components in self-assembling systems =====
-    * [[http://code.ulb.ac.be/iridia.people.php?id=1|Hugues Bersini (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~stuetzle/|Thomas Stützle (IRIDIA)]]
+The goal of this project is to apply computer vision techniques to track the growth of structures in self-assembling systems. The ability to track the growth of structures will shed light on the dynamics of self-assembly; an aspect of self-assembly that has not been well researched in the macroscopic context.
+As part of the project, the student will have to: (a) find suitable hardware (combination of camera, lens, lighting, etc.), (b) localise multiple components in an environment, (c) track the components in the environment, and (d) track the assembly of components in the environment.
+Required skills: the candidates should be able to program in C++ and have a working knowledge of the English language.
+ * Contacts : [[http://iridia.ulb.ac.be/~mdorigo/HomePageDorigo/ | Marco Dorigo]] and Dhananjay Ipparthi (IRIDIA)
@@ Line 379: / Line 496: @@
-===== Comparison of fast heuristics for the longest common subsequence problem =====
-The [[http://en.wikipedia.org/wiki/Longest_common_subsequence|longest common subsequence (LCS) problem]] has important applications in Computational Biology. Several heuristic methods have been proposed to obtain approximate solutions. These methods require different computation time and obtain solutions of varied quality. In this project, the student will learn several methods that have been proposed in the literature to tackle a difficult optimization problem, and compare them in terms of computation time and quality of the resulting solutions. The final goal is to propose appropriate combinations of existing methods that solve diverse instances of the LCS problem.
+===== Applications of a hybrid SLS algorithm framework =====
-  * Contacts :
+We have recently developed a software framework from which hybrid stochastic local search algorithms can be designed automatically. This framework has only been applied to a few problems. The goal of this project would be to extend this framework to other problems and compare its results with the methods proposed in the literature. The student will learn to solve combinatorial optimization problems with SLS algorithms, automatic configuration of optimization algorithms, and analysis and comparison of optimization algorithms.
-    * [[http://iridia.ulb.ac.be/~manuel|Manuel López-Ibáñez (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stuetzle (IRIDIA)]]
-===== Applications of the Multi-objective ACO framework =====
-We have recently developed a software framework of Ant Colony Optimization algorithms for multi-objective optimization problems. This framework has only been applied to a few problems. The goal of this project would be to extend this framework to other problems and compare its results with the methods proposed in the literature. The student will learn to solve multi-objective optimization problems with ACO algorithms, automatic configuration of optimization algorithms, and analysis and comparison of optimization algorithms for multi-objective problems.
   * Contacts :
-    * [[http://iridia.ulb.ac.be/~manuel|Manuel López-Ibáñez (IRIDIA)]]
+    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stützle (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stuetzle (IRIDIA)]]
+    * [[http://code.ulb.ac.be/iridia.people.php?id=1388|Federico Pagnozzi (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~mdorigo|Marco Dorigo (IRIDIA)]]
-===== A graphical interface for the optimisation of Water Distribution Networks =====
-The [[http://iridia.ulb.ac.be/~manuel/doc/cec2005-presentation.pdf|optimization of the operations of Water Distribution Networks]] may save important amounts of energy and its associated costs, and, therefore, it is an important problem in practice. There are [[http://www.epa.gov/nrmrl/wswrd/dw/epanet.html|graphical tools and simulators]] available. In addition, several optimization methods based on [[http://iridia.ulb.ac.be/~manuel/doc/cec2005.pdf|evolutionary algorithms]] and [[http://dx.doi.org/10.1061/(ASCE)0733-9496(2008)134:4(337)|ant colony optimization]] have been proposed in the literature. The goal of this project is to integrate the optimization algorithms into a graphical environment that can be used by water engineers and operators. No knowledge about water distribution networks is necessary. The optimisation algorithms and toolkit libraries for handling water distribution networks will be available to the student.
-  * Contacts :
-    * [[http://iridia.ulb.ac.be/~manuel|Manuel López-Ibáñez (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stuetzle (IRIDIA)]]
-===== Automatic fine-tuning of an evolutionary multi-objective framework =====
-The goal of this project is to explore the possibilities of using automatic configuration tools for fine-tuning an existing [[http://paradiseo.gforge.inria.fr/index.php?n=Paradiseo.MOEO|evolutionary multi-objective framework]]. The student will learn about automatic configuration tools, evolutionary algorithms for multi-objective optimization problems and analysis and comparison of multi-objective algorithms.
-  * Contacts :
-    * [[http://iridia.ulb.ac.be/~manuel|Manuel López-Ibáñez (IRIDIA)]]
-    * [[http://iridia.ulb.ac.be/~stuetzle|Thomas Stuetzle (IRIDIA)]]