Browse ORBi by ORBi project

- Background
- Content
- Benefits and challenges
- Legal aspects
- Functions and services
- Team
- Help and tutorials

Tree based ensemble models regularization by convex optimization Cornélusse, Bertrand ; Geurts, Pierre ; Wehenkel, Louis Conference (2009, December 12) Tree based ensemble methods can be seen as a way to learn a kernel from a sample of input-output pairs. This paper proposes a regularization framework to incorporate non-standard information not used in ... [more ▼] Tree based ensemble methods can be seen as a way to learn a kernel from a sample of input-output pairs. This paper proposes a regularization framework to incorporate non-standard information not used in the kernel learning algorithm, so as to take advantage of incomplete information about output values and/or of some prior information about the problem at hand. To this end a generic convex optimization problem is formulated which is first customized into a manifold regularization approach for semi-supervised learning, then as a way to exploit censored output values, and finally as a generic way to exploit prior information about the problem. [less ▲] Detailed reference viewed: 140 (45 ULg)Large Margin Classification with the Progressive Hedging Algorithm Defourny, Boris ; Wehenkel, Louis Conference (2009, December) Several learning algorithms in classification and structured prediction are formulated as large scale optimization problems. We show that a generic iterative reformulation and resolving strategy based on ... [more ▼] Several learning algorithms in classification and structured prediction are formulated as large scale optimization problems. We show that a generic iterative reformulation and resolving strategy based on the progressive hedging algorithm from stochastic programming results in a highly parallel algorithm when applied to the large margin classification problem with nonlinear kernels. We also underline promising aspects of the available analysis of progressive hedging strategies. [less ▲] Detailed reference viewed: 53 (4 ULg)Supervised learning with decision tree-based methods in computational and systems biology Geurts, Pierre ; Irrthum, Alexandre ; Wehenkel, Louis in Molecular Biosystems (2009), 5(12), 1593-1605 At the intersection between artiﬁcial intelligence and statistics, supervised learning provides algorithms to automatically build predictive models only from observations of a system. During the last ... [more ▼] At the intersection between artiﬁcial intelligence and statistics, supervised learning provides algorithms to automatically build predictive models only from observations of a system. During the last twenty years, supervised learning has been a tool of choice to analyze the always increasing and complexifying data generated in the context of molecular biology, with successful applications in genome annotation, function prediction, or biomarker discovery. Among supervised learning methods, decision tree-based methods stand out as non parametric methods that have the unique feature of combining interpretability, eﬃciency, and, when used in ensembles of trees, excellent accuracy. The goal of this paper is to provide an accessible and comprehensive introduction to this class of methods. The ﬁrst part of the paper is devoted to an intuitive but complete description of decision tree-based methods and a discussion of their strengths and limitations with respect to other supervised learning methods. The second part of the paper provides a survey of their applications in the context of computational and systems biology. The supplementary material provides information about various non-standard extensions of the decision tree-based approach to modeling, some practical guidelines for the choice of parameters and algorithm variants depending on the practical ob jectives of their application, pointers to freely accessible software packages, and a brief primer going through the diﬀerent manipulations needed to use the tree-induction packages available in the R statistical tool. [less ▲] Detailed reference viewed: 183 (31 ULg)Inferring regulatory networks from expression data using tree-based methods Huynh-Thu, Vân Anh ; Irrthum, Alexandre ; Wehenkel, Louis et al Poster (2009, September) Detailed reference viewed: 20 (11 ULg)Security Assessment by Multiple Transmission System Operators Exchanging Sensitivity and Tie-Line Power Flow Information Fabozzi, Davide ; ; Wehenkel, Louis et al in Proceedings of the 2009 IEEE PES Power Tech conference (2009, July) This paper considers a procedure for multi-area static security assessment of large interconnected power systems operated by a team of Transmission System Operators (TSOs). In this procedure, each TSO ... [more ▼] This paper considers a procedure for multi-area static security assessment of large interconnected power systems operated by a team of Transmission System Operators (TSOs). In this procedure, each TSO provides the other TSOs with his own equivalent model as well as the detailed effects of contingencies in his control area on all tie-line flows. The paper deals with the implementation of sensitivity-based equivalents suitable for static security assessment. Accuracy with respect to the unreduced model and computational efficiency are considered in evaluating the proposed approach. The relevance of the procedure in the context of recent UCTE operational security policy recommendations is also stressed. The procedure has been implemented in an AC power flow program and tested on a three-area variant of the IEEE 118-bus test system. [less ▲] Detailed reference viewed: 108 (23 ULg)A new heuristic approach to deal with discrete variables in optimal power flow computations Capitanescu, Florin ; Wehenkel, Louis in IEEE Power Tech conference (2009, July) This paper proposes a new heuristic approach to deal with discrete variables in an optimal power flow (OPF). This approach relies on the first order sensitivity of the objective and inequality constraints ... [more ▼] This paper proposes a new heuristic approach to deal with discrete variables in an optimal power flow (OPF). This approach relies on the first order sensitivity of the objective and inequality constraints with respect to the discrete variables. The impact of a discrete variable change on the objective and inequality constraints is aggregated into a merit function. The proposed approach searches iteratively for better discrete variable settings as long as the problem solution can be improved. We provide numerical results with the proposed approach on four test systems up to 1203 buses and for the OPF problem of active power loss minimization. [less ▲] Detailed reference viewed: 67 (9 ULg)Optimal power flow computations with constraints limiting the number of control actions Capitanescu, Florin ; ; Wehenkel, Louis in IEEE Power Tech conference (2009, July) This paper focuses on optimal power flow (OPF) computations in which no more than a pre-specified number of controls are allowed to move. The benchmark formulation of this OPF problem constitutes a mixed ... [more ▼] This paper focuses on optimal power flow (OPF) computations in which no more than a pre-specified number of controls are allowed to move. The benchmark formulation of this OPF problem constitutes a mixed integer nonlinear programming (MINLP) problem. To avoid the prohibitive computational time required by classical MINLP approaches to provide a (potentially sub-optimal) solution, we propose instead two alternative approaches. The first one consists in reformulating the MINLP problem as a mathematical program with equilibrium constraints (MPEC). The second approach includes in the classical OPF problem a nonlinear constraint which approximates the integral constraint limiting the number of control variables movement. Both approaches are solved by an interior point algorithm (IPA), slightly adapted to the particular characteristics of each approach. We provide numerical results with the proposed approaches on two test systems and for two practical problems: minimum cost to remove thermal congestion, and minimum cost of load curtailment to restore a feasible equilibrium point. [less ▲] Detailed reference viewed: 34 (2 ULg)Coupling optimization and dynamic simulation for preventive-corrective control of voltage instability Capitanescu, Florin ; Van Cutsem, Thierry ; Wehenkel, Louis in IEEE Transactions on Power Systems (2009), 24(2), 796-805 This paper proposes an approach coupling security constrained optimal power flow with time-domain simulation to determine an optimal combination of preventive and corrective controls ensuring a voltage ... [more ▼] This paper proposes an approach coupling security constrained optimal power flow with time-domain simulation to determine an optimal combination of preventive and corrective controls ensuring a voltage stable transition of the system towards a feasible long-term equilibrium, if any of a set of postulated contingencies occurs. A security-constrained optimal power flow is used to adjust the respective contribution of preventive and corrective actions. Furthermore, information is extracted from (quasi steady-state) time-domain simulations to iteratively adjust the set of coupling constraints used by a corrective security constrained optimal power flow until its solution is found dynamically secure and viable. Numerical results are provided on a realistic 55-bus test system. [less ▲] Detailed reference viewed: 118 (19 ULg)Fast Multi-Class Image Annotation with Random Subwindows and Multiple Output Randomized Trees ; Marée, Raphaël ; Wehenkel, Louis et al in Proc. International Conference on Computer Vision Theory and Applications (VISAPP) (2009, February) This paper addresses image annotation, i.e. labelling pixels of an image with a class among a finite set of predefined classes. We propose a new method which extracts a sample of subwindows from a set of ... [more ▼] This paper addresses image annotation, i.e. labelling pixels of an image with a class among a finite set of predefined classes. We propose a new method which extracts a sample of subwindows from a set of annotated images in order to train a subwindow annotation model by using the extremely randomized trees ensemble method appropriately extended to handle high-dimensional output spaces. The annotation of a pixel of an unseen image is done by aggregating the annotations of its subwindows containing this pixel. The proposed method is compared to a more basic approach predicting the class of a pixel from a single window centered on that pixel and to other state-of-the-art image annotation methods. In terms of accuracy, the proposed method significantly outperforms the basic method and shows good performances with respect to the state-of-the-art, while being more generic, conceptually simpler, and of higher computational efficiency than these latter. [less ▲] Detailed reference viewed: 191 (21 ULg)Content-based Image Retrieval by Indexing Random Subwindows with Randomized Trees Marée, Raphaël ; Geurts, Pierre ; Wehenkel, Louis in IPSJ Transactions on Computer Vision and Applications (2009), 1 We propose a new method for content-based image retrieval which exploits the similarity measure and indexing structure of totally randomized tree ensembles induced from a set of subwindows randomly ... [more ▼] We propose a new method for content-based image retrieval which exploits the similarity measure and indexing structure of totally randomized tree ensembles induced from a set of subwindows randomly extracted from a sample of images. We also present the possibility of updating the model as new images come in, and the capability of comparing new images using a model previously constructed from a different set of images. The approach is quantitatively evaluated on various types of images and achieves high recognition rates despite its conceptual simplicity and computational efficiency. [less ▲] Detailed reference viewed: 157 (25 ULg)What is the likely future of real-time transient stability ? Ernst, Damien ; Wehenkel, Louis ; Pavella, Mania in Proceedings of the 2009 IEEE/PES Power Systems Conference & Exposition (PSCE 2009) (2009) Despite very intensive research efforts in the field of transient stability during the last five decades, the large majority of the derived techniques have hardly moved from the research laboratories to ... [more ▼] Despite very intensive research efforts in the field of transient stability during the last five decades, the large majority of the derived techniques have hardly moved from the research laboratories to the industrial world and, as a matter of fact, the very large majority of today's control centers do not make use of any real-time transient stability software. On the other hand, along all these years the techniques developed for real-time transient stability have mainly focused on the definition of stability margins and speeding-up techniques rather than on preventive or emergency control strategies. In the light of the above observations, this paper attempts to explain the reasons for lack of industrial interest in real-time transient stability, and also to examine an even more fundamental question, namely: is transient stability, as has been stated many decades ago, still the relevant issue in the context of the new power systems morphology towards more dispersed generation, higher penetration of power electronics, larger and more complex structures, and, in addition, of economic and environmental constraints? Or, maybe, there is a need for techniques different from those developed so far? [less ▲] Detailed reference viewed: 133 (19 ULg)Inferring bounds on the performance of a control policy from a sample of one-step system transitions Fonteneau, Raphaël ; ; Wehenkel, Louis et al in 28th Benelux Meeting on Systems and Control (2009) Detailed reference viewed: 12 (4 ULg)Dynamic treatment regimes using reinforcement learning: a cautious generalization approach Fonteneau, Raphaël ; ; Wehenkel, Louis et al Poster (2009) Detailed reference viewed: 9 (1 ULg)Biomarker discovery in asthma-related inflammation and remodeling. Quesada Calvo, Florence ; Fillet, Marianne ; De Seny, Dominique et al in Proteomics (2009), 9(8), 2163-2170 Asthma is a complex inflammatory disease of airways. A network of reciprocal interactions between inflammatory cells, peptidic mediators, extracellular matrix components, and proteases is thought to be ... [more ▼] Asthma is a complex inflammatory disease of airways. A network of reciprocal interactions between inflammatory cells, peptidic mediators, extracellular matrix components, and proteases is thought to be involved in the installation and maintenance of asthma-related airway inflammation and remodeling. To date, new proteic mediators displaying significant activity in the pathophysiology of asthma are still to be unveiled. The main objective of this study was to uncover potential target proteins by using surface-enhanced laser desorption/ionization-time of flight-mass spectrometry (SELDI-TOF-MS) on lung samples from mouse models of allergen-induced airway inflammation and remodeling. In this model, we pointed out several protein or peptide peaks that were preferentially expressed in diseased mice as compared to controls. We report the identification of different five proteins: found inflammatory zone 1 or RELM (FIZZ-1), calcyclin (S100A6), clara cell secretory protein 10 (CC10), Ubiquitin, and Histone H4. [less ▲] Detailed reference viewed: 276 (76 ULg)Inferring regulatory networks from expression data using tree-based methods Huynh-Thu, Vân Anh ; Irrthum, Alexandre ; Wehenkel, Louis et al Conference (2009) Detailed reference viewed: 8 (2 ULg)Protéomique par SELDI-TOF-MS des maladies inflammatoires articulaires: identification des protéines S100 comme protéines d'intérêt De Seny, Dominique ; Ribbens, Clio ; Cobraiville, Gaël et al in Revue Médicale de Liège (2009), 64(Spec No), 29-35 Clinical proteomics is a technical approach studying the entire proteome expressed by cells, tissues or organs. It describes the dynamics of cell regulation by detecting molecular events related to ... [more ▼] Clinical proteomics is a technical approach studying the entire proteome expressed by cells, tissues or organs. It describes the dynamics of cell regulation by detecting molecular events related to diseases development. Proteomic techniques focus mainly on identification of new biomarkers or new therapeutic targets. It is a multidisciplinary approach using medical, biological, bioanalytical and bioinformatics knowledges. A strong collaboration between these fields allowed SELDI-TOF-MS proteomics studies to be performed at the CHU and the University of Liege, in GIGA-Research facilities. The aim of these studies was driven along three main axes of research related to the identification of biomarkers specific to a studied pathology, to a common biological pathway and, finally, to a treatment response. [less ▲] Detailed reference viewed: 88 (8 ULg)An Extra-trees-based Automatic Target Recognition Algorithm Pisane, Jonathan ; Marée, Raphaël ; Ries, Philippe et al in To appear in Proc. International Radar Conference (2009) Detailed reference viewed: 120 (33 ULg)Inferring bounds on the performance of a control policy from a sample of trajectories Fonteneau, Raphaël ; ; Wehenkel, Louis et al in Proceedings of the IEEE International Symposium on Adaptive Dynamic Programming and Reinforcement Learning (ADPRL-09) (2009) We propose an approach for inferring bounds on the finite-horizon return of a control policy from an off-policy sample of trajectories collecting state transitions, rewards, and control actions. In this ... [more ▼] We propose an approach for inferring bounds on the finite-horizon return of a control policy from an off-policy sample of trajectories collecting state transitions, rewards, and control actions. In this paper, the dynamics, control policy, and reward function are supposed to be deterministic and Lipschitz continuous. Under these assumptions, a polynomial algorithm, in terms of the sample size and length of the optimization horizon, is derived to compute these bounds, and their tightness is characterized in terms of the sample density. [less ▲] Detailed reference viewed: 37 (10 ULg)Planning under uncertainty, ensembles of disturbance trees and kernelized discrete action spaces Defourny, Boris ; Ernst, Damien ; Wehenkel, Louis in Proceedings of the IEEE International Symposium on Adaptive Dynamic Programming and Reinforcement Learning (ADPRL-09) (2009) Optimizing decisions on an ensemble of incomplete disturbance trees and aggregating their first stage decisions has been shown as a promising approach to (model-based) planning under uncertainty in large ... [more ▼] Optimizing decisions on an ensemble of incomplete disturbance trees and aggregating their first stage decisions has been shown as a promising approach to (model-based) planning under uncertainty in large continuous action spaces and in small discrete ones. The present paper extends this approach and deals with large but highly structured action spaces, through a kernel-based aggregation scheme. The technique is applied to a test problem with a discrete action space of 6561 elements adapted from the NIPS 2005 SensorNetwork benchmark. [less ▲] Detailed reference viewed: 30 (10 ULg)A rare-event approach to build security analysis tools when N-k (k > 1) analyses are needed (as they are in large-scale power systems) Belmudes, Florence ; Ernst, Damien ; Wehenkel, Louis in Proceedings of the 2009 IEEE Bucharest PowerTech (2009) We consider the problem of performing N − k security analyses in large scale power systems. In such a context, the number of potentially dangerous N − k contingencies may become rapidly very large when k ... [more ▼] We consider the problem of performing N − k security analyses in large scale power systems. In such a context, the number of potentially dangerous N − k contingencies may become rapidly very large when k grows, and so running a security analysis for each one of them is often intractable. We assume in this paper that the number of dangerous N − k contingencies is very small with respect to the number of non-dangerous ones. Under this assumption, we suggest to use importance sampling techniques for identifying rare events in combinatorial search spaces. With such techniques, it is possible to identify dangerous contingencies by running security analyses for only a small number of events. A procedure relying on these techniques is proposed in this work for steady-state security analyses. This procedure has been evaluated on the IEEE 118 bus test system. The results show that it is indeed able to efficiently identify among a large set of contingencies some of the rare ones which are dangerous. [less ▲] Detailed reference viewed: 49 (7 ULg) |
||