References of "Wehenkel, Louis"
     in
Bookmark and Share    
Full Text
Peer Reviewed
See detailOptimal sample selection for batch-mode reinforcement learning
Rachelson, Emmanuel ULg; Schnitzler, François ULg; Wehenkel, Louis ULg et al

in Proceedings of the 3rd International Conference on Agents and Artificial Intelligence (ICAART 2011) (2011)

We introduce the Optimal Sample Selection (OSS) meta-algorithm for solving discrete-time Optimal Control problems. This meta-algorithm maps the problem of finding a near-optimal closed-loop policy to the ... [more ▼]

We introduce the Optimal Sample Selection (OSS) meta-algorithm for solving discrete-time Optimal Control problems. This meta-algorithm maps the problem of finding a near-optimal closed-loop policy to the identification of a small set of one-step system transitions, leading to high-quality policies when used as input of a batch-mode Reinforcement Learning (RL) algorithm. We detail a particular instance of this OSS metaalgorithm that uses tree-based Fitted Q-Iteration as a batch-mode RL algorithm and Cross Entropy search as a method for navigating efficiently in the space of sample sets. The results show that this particular instance of OSS algorithms is able to identify rapidly small sample sets leading to high-quality policies [less ▲]

Detailed reference viewed: 85 (13 ULg)
Full Text
See detailMultistage stochastic programming: A scenario tree based approach to planning under uncertainty
Defourny, Boris ULg; Ernst, Damien ULg; Wehenkel, Louis ULg

in Sucar, L. Enrique; Morales, Eduardo F.; Hoey, Jesse (Eds.) Decision Theory Models for Applications in Artificial Intelligence: Concepts and Solutions (2011)

In this chapter, we present the multistage stochastic programming framework for sequential decision making under uncertainty. We discuss its differences with Markov Decision Processes, from the point of ... [more ▼]

In this chapter, we present the multistage stochastic programming framework for sequential decision making under uncertainty. We discuss its differences with Markov Decision Processes, from the point of view of decision models and solution algorithms. We describe the standard technique for solving approximately multistage stochastic problems, which is based on a discretization of the disturbance space called scenario tree. We insist on a critical issue of the approach: the decisions can be very sensitive to the parameters of the scenario tree, whereas no efficient tool for checking the quality of approximate solutions exists. In this chapter, we show how supervised learning techniques can be used to evaluate reliably the quality of an approximation, and thus facilitate the selection of a good scenario tree. The framework and solution techniques presented in the chapter are explained and detailed on several examples. Along the way, we define notions from decision theory that can be used to quantify, for a particular problem, the advantage of switching to a more sophisticated decision model. [less ▲]

Detailed reference viewed: 226 (31 ULg)
Full Text
Peer Reviewed
See detailSensitivity-based approaches for handling discrete variables in optimal power flow computations
Capitanescu, Florin ULg; Wehenkel, Louis ULg

in IEEE Transactions on Power Systems (2010), 25(4), 1780-1789

This paper proposes and compares three iterative approaches for handling discrete variables in optimal power flow (OPF) computations. The first two approaches rely on the sensitivities of the objective ... [more ▼]

This paper proposes and compares three iterative approaches for handling discrete variables in optimal power flow (OPF) computations. The first two approaches rely on the sensitivities of the objective and inequality constraints with respect to discrete variables. They set the discrete variables values either by solving a mixed-integer linear programming (MILP) problem or by using a simple procedure based on a merit function. The third approach relies on the use of Lagrange multipliers corresponding to the discrete variables bound constraints at the OPF solution. The classical round-off technique and a progressive round-off approach have been also used as a basis of comparison. We provide extensive numerical results with these approaches on four test systems with up to 1203 buses, and for two OPF problems: loss minimization and generation cost minimization, respectively. These results show that the sensitivity-based approach combined with the merit function clearly outperforms the other approaches in terms of: objective function quality, reliability, and computational times. Furthermore, the objective value obtained with this approach has been very close to that provided by the continuous relaxation OPF. This approach constitutes therefore a viable alternative to other methods dealing with discrete variables in an OPF. [less ▲]

Detailed reference viewed: 52 (4 ULg)
Full Text
Peer Reviewed
See detailInferring Regulatory Networks from Expression Data Using Tree-Based Methods
Huynh-Thu, Vân Anh ULg; Irrthum, Alexandre ULg; Wehenkel, Louis ULg et al

in PLoS ONE (2010), 5(9), 12776

One of the pressing open problems of computational systems biology is the elucidation of the topology of genetic regulatory networks (GRNs) using high throughput genomic data, in particular microarray ... [more ▼]

One of the pressing open problems of computational systems biology is the elucidation of the topology of genetic regulatory networks (GRNs) using high throughput genomic data, in particular microarray gene expression data. The Dialogue for Reverse Engineering Assessments and Methods (DREAM) challenge aims to evaluate the success of GRN inference algorithms on benchmarks of simulated data. In this article, we present GENIE3, a new algorithm for the inference of GRNs that was best performer in the DREAM4 In Silico Multifactorial challenge. GENIE3 decomposes the prediction of a regulatory network between p genes into p different regression problems. In each of the regression problems, the expression pattern of one of the genes (target gene) is predicted from the expression patterns of all the other genes (input genes), using tree-based ensemble methods Random Forests or Extra-Trees. The importance of an input gene in the prediction of the target gene expression pattern is taken as an indication of a putative regulatory link. Putative regulatory links are then aggregated over all genes to provide a ranking of interactions from which the whole network is reconstructed. In addition to performing well on the DREAM4 In Silico Multifactorial challenge simulated data, we show that GENIE3 compares favorably with existing algorithms to decipher the genetic regulatory network of Escherichia coli. It doesn't make any assumption about the nature of gene regulation, can deal with combinatorial and non-linear interactions, produces directed GRNs, and is fast and scalable. In conclusion, we propose a new algorithm for GRN inference that performs well on both synthetic and real gene expression data. The algorithm, based on feature selection with tree-based ensemble methods, is simple and generic, making it adaptable to other types of genomic data and interactions. [less ▲]

Detailed reference viewed: 364 (40 ULg)
Full Text
Peer Reviewed
See detailSub-quadratic Markov tree mixture learning based on randomizations of the Chow-Liu algorithm
Ammar, Sourour; Leray, Philippe; Schnitzler, François ULg et al

in Myllymäki, Petri; Roos, Antoine; Jaakkola, Tommi (Eds.) Proceedings of the Fifth European Workshop on Probabilistic Graphical Models (PGM-2010) (2010, September)

The present work analyzes different randomized methods to learn Markov tree mixtures for density estimation in very high-dimensional discrete spaces (very large number n of discrete variables) when the ... [more ▼]

The present work analyzes different randomized methods to learn Markov tree mixtures for density estimation in very high-dimensional discrete spaces (very large number n of discrete variables) when the sample size (N ) is very small compared to n. Several sub- quadratic relaxations of the Chow-Liu algorithm are proposed, weakening its search proce- dure. We first study na¨ıve randomizations and then gradually increase the deterministic behavior of the algorithms by trying to focus on the most interesting edges, either by retaining the best edges between models, or by inferring promising relationships between variables. We compare these methods to totally random tree generation and randomiza- tion based on bootstrap-resampling (bagging), of respectively linear and quadratic com- plexity. Our results show that randomization becomes increasingly more interesting for smaller N/n ratios, and that methods based on simultaneously discovering and exploiting the problem structure are promising in this context. [less ▲]

Detailed reference viewed: 36 (19 ULg)
Full Text
Peer Reviewed
See detailConsequence driven decomposition of large-scale power system security analysis
Fonteneau, Florence ULg; Ernst, Damien ULg; Druet, Christophe et al

in Proceedings of the 2010 IREP Symposium - Bulk Power Systems Dynamics and Control - VIII (2010, August)

This paper presents an approach for assessing, in operation planning studies, the security of a large-scale power system by decomposing it into elementary subproblems, each one corresponding to a ... [more ▼]

This paper presents an approach for assessing, in operation planning studies, the security of a large-scale power system by decomposing it into elementary subproblems, each one corresponding to a structural weak-point of the system. We suppose that the structural weak-points are known a priori by the system operators, and are each one described by a set of constraints that are localized in some relatively small area of the system. The security analysis with respect to a given weak-point thus reduces to the identification of the combinations of power system configurations and disturbances that could lead to the violation of some of its constraints. We propose an iterative rare-event simulation approach for identifying such combinations among the very large set of possible ones. The procedure is illustrated on a simplified version of this problem applied to the Belgian transmission system. [less ▲]

Detailed reference viewed: 53 (11 ULg)
Peer Reviewed
See detailRadar Classification based on Extra-Trees
Pisane, Jonathan ULg; Marée, Raphaël ULg; Wehenkel, Louis ULg et al

(2010, May 24)

In this paper, we describe a new automatic target recognition algorithm for classifying SAR images based on the PiXiT im- age classifier. It uses randomized sub-windows extraction and extremely randomized ... [more ▼]

In this paper, we describe a new automatic target recognition algorithm for classifying SAR images based on the PiXiT im- age classifier. It uses randomized sub-windows extraction and extremely randomized trees (extra-trees). This approach re- quires very little pre-processing of the images, thereby lim- iting the computational load. It was successfully tested on an extended version of the public standard MSTAR database, that includes targets of interest, false targets, and background clutter. A misclassification rate of about three percent has been achieved. [less ▲]

Detailed reference viewed: 58 (7 ULg)
Full Text
Peer Reviewed
See detailVers un apprentissage subquadratique pour les mélanges d’arbres
Schnitzler, François ULg; Leray, Philippe; Wehenkel, Louis ULg

Conference (2010, May 10)

We consider randomization schemes of the Chow-Liu algorithm from weak (bagging, of quadratic complexity) to strong ones (full random sampling, of linear complexity), for learn- ing probability density ... [more ▼]

We consider randomization schemes of the Chow-Liu algorithm from weak (bagging, of quadratic complexity) to strong ones (full random sampling, of linear complexity), for learn- ing probability density models in the form of mixtures of Markov trees. Our empirical study on high-dimensional synthetic problems shows that, while bagging is the most accurate scheme on average, some of the stronger randomizations remain very competitive in terms of accuracy, specially for small sample sizes. [less ▲]

Detailed reference viewed: 44 (16 ULg)
Full Text
Peer Reviewed
See detailModel-free Monte Carlo–like policy evaluation
Fonteneau, Raphaël ULg; Murphy, Susan; Wehenkel, Louis ULg et al

in Proceedings of Conférence Francophone sur l'Apprentissage Automatique (CAp) 2010 (2010, May)

We propose an algorithm for estimating the finite-horizon expected return of a closed loop control policy from an a priori given (off-policy) sample of one-step transitions. It averages cumulated rewards ... [more ▼]

We propose an algorithm for estimating the finite-horizon expected return of a closed loop control policy from an a priori given (off-policy) sample of one-step transitions. It averages cumulated rewards along a set of “broken trajectories” made of one-step transitions selected from the sample on the basis of the control policy. Under some Lipschitz continuity assumptions on the system dynamics, reward function and control policy, we provide bounds on the bias and variance of the estimator that depend only on the Lipschitz constants, on the number of broken trajectories used in the estimator, and on the sparsity of the sample of one-step transitions. [less ▲]

Detailed reference viewed: 23 (10 ULg)
Full Text
Peer Reviewed
See detailModel-free Monte Carlo-like policy evaluation
Fonteneau, Raphaël ULg; Murphy, Susan; Wehenkel, Louis ULg et al

in Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics (AISTATS 2010) (2010, May)

We propose an algorithm for estimating the finite-horizon expected return of a closed loop control policy from an a priori given (off-policy) sample of one-step transitions. It averages cumulated rewards ... [more ▼]

We propose an algorithm for estimating the finite-horizon expected return of a closed loop control policy from an a priori given (off-policy) sample of one-step transitions. It averages cumulated rewards along a set of “broken trajectories” made of one-step transitions selected from the sample on the basis of the control policy. Under some Lipschitz continuity assumptions on the system dynamics, reward function and control policy, we provide bounds on the bias and variance of the estimator that depend only on the Lipschitz constants, on the number of broken trajectories used in the estimator, and on the sparsity of the sample of one-step transitions. [less ▲]

Detailed reference viewed: 68 (16 ULg)
Full Text
Peer Reviewed
See detailGenerating informative trajectories by using bounds on the return of control policies
Fonteneau, Raphaël ULg; Murphy, Susan; Wehenkel, Louis ULg et al

in Proceedings of the Workshop on Active Learning and Experimental Design 2010 (in conjunction with AISTATS 2010) (2010, May)

We propose new methods for guiding the generation of informative trajectories when solving discrete-time optimal control problems. These methods exploit recently published results that provide ways for ... [more ▼]

We propose new methods for guiding the generation of informative trajectories when solving discrete-time optimal control problems. These methods exploit recently published results that provide ways for computing bounds on the return of control policies from a set of trajectories. [less ▲]

Detailed reference viewed: 33 (12 ULg)
Full Text
Peer Reviewed
See detailTowards sub-quadratic learning of probability density models in the form of mixtures of trees
Schnitzler, François ULg; Leray, Philippe; Wehenkel, Louis ULg

(2010, April)

We consider randomization schemes of the Chow-Liu algorithm from weak (bagging, of quadratic complexity) to strong ones (full random sampling, of linear complexity), for learning probability density ... [more ▼]

We consider randomization schemes of the Chow-Liu algorithm from weak (bagging, of quadratic complexity) to strong ones (full random sampling, of linear complexity), for learning probability density models in the form of mixtures of Markov trees. Our empirical study on high-dimensional synthetic problems shows that, while bagging is the most accurate scheme on average, some of the stronger randomizations remain very competitive in terms of accuracy, specially for small sample sizes. [less ▲]

Detailed reference viewed: 58 (21 ULg)
Peer Reviewed
See detailZebrafish as model in toxicology/pharmacology.
Voncken, Audrey ULg; Piot, Amandine ULg; Stern, Olivier ULg et al

Poster (2010, March 17)

Detailed reference viewed: 76 (30 ULg)
Full Text
Peer Reviewed
See detailIncremental Indexing and Distributed Image Search using Shared Randomized Vocabularies
Marée, Raphaël ULg; Denis, Philippe; Wehenkel, Louis ULg et al

in ACM Proceedings MIR 2010 (2010, March)

We present a cooperative framework for content-based image retrieval for the realistic setting where images are distributed across multiple cooperating servers. The proposed method is in line with bag-of ... [more ▼]

We present a cooperative framework for content-based image retrieval for the realistic setting where images are distributed across multiple cooperating servers. The proposed method is in line with bag-of-features approaches but uses fully data-independent, randomized structures, shared by the cooperating servers, to map image features to common visual words. A coherent, global image similarity measure (which is a kernel) is computed in a distributed fashion over visual words, by only requiring a small amount of data transfers between nodes. Our experiments on various image types show that this framework is a very promising step towards large-scale, distributed content-based image retrieval. [less ▲]

Detailed reference viewed: 88 (13 ULg)
Full Text
Peer Reviewed
See detailOptimal power flow computations with a limited number of controls allowed to move
Capitanescu, Florin ULg; Wehenkel, Louis ULg

in IEEE Transactions on Power Systems (2010), 25(1), 586-587

This letter focuses on optimal power flow (OPF) computations in which no more than a pre-specified number of controls are allowed to move. To determine an efficient subset of controls satisfying this ... [more ▼]

This letter focuses on optimal power flow (OPF) computations in which no more than a pre-specified number of controls are allowed to move. To determine an efficient subset of controls satisfying this constraint we rely on the solution of a mixed integer linear programming (MILP) problem fed with sensitivity information of controls' impact on the objective and constraints. We illustrate this approach on a 60-bus system and for the OPF problem of minimum load curtailment cost to remove thermal congestion. [less ▲]

Detailed reference viewed: 62 (7 ULg)
Full Text
Peer Reviewed
See detailA cautious approach to generalization in reinforcement learning
Fonteneau, Raphaël ULg; Murphy, Susan; Wehenkel, Louis ULg et al

in Proceedings of the 2nd International Conference on Agents and Artificial Intelligence (2010, January)

In the context of a deterministic Lipschitz continuous environment over continuous state spaces, finite action spaces, and a finite optimization horizon, we propose an algorithm of polynomial complexity ... [more ▼]

In the context of a deterministic Lipschitz continuous environment over continuous state spaces, finite action spaces, and a finite optimization horizon, we propose an algorithm of polynomial complexity which exploits weak prior knowledge about its environment for computing from a given sample of trajectories and for a given initial state a sequence of actions. The proposed Viterbi-like algorithm maximizes a recently proposed lower bound on the return depending on the initial state, and uses to this end prior knowledge about the environment provided in the form of upper bounds on its Lipschitz constants. It thereby avoids, in way depending on the initial state and on the prior knowledge, those regions of the state space where the sample is too sparse to make safe generalizations. Our experiments show that it can lead to more cautious policies than algorithms combining dynamic programming with function approximators. We give also a condition on the sample sparsity ensuring that, for a given initial state, the proposed algorithm produces an optimal sequence of actions in open-loop. [less ▲]

Detailed reference viewed: 84 (22 ULg)
Full Text
Peer Reviewed
See detailModel-free Monte Carlo-like policy evaluation
Fonteneau, Raphaël ULg; Murphy, Susan A.; Wehenkel, Louis ULg et al

in 29th Benelux Meeting on Systems and Control (2010)

Detailed reference viewed: 9 (0 ULg)