National audienceEfficient search of (quasi-)optimal paths in graphs remains a fundamental task in Artificial Intelligence. Recent works [7, 4, 5, 8] have contributed to a new point of view on this problem whereby heuristics are learned from past solving experiences rather than derived through a static abstraction of the description of the problem. In this paper, we show how to improve this work by better exploiting information from past solving episodes. The experiments reported here confirm the significant reduction in search space achieved by our algorithm. In a second part, we show how to generalize these learning techniques to the case of changing goal states. Extensive experiments and their analysis show that the variations of the goa...
National audienceDans cet article, nous présenterons les travaux prélimi-naires menés sur l'utilisat...
The impressive breakthroughs of the last two decades in the field of machine learning can be in larg...
L'apprentissage structuré est au fondement des méthodes modernes d'apprentissage automatique pour le...
La recherche efficace d'un chemin (quasi) optimal dans un graphe reste une tâche fondamentale en Inte...
The last couple of decades have seen a surge of interest and sophistication in using heuristics to s...
This thesis presents our contributions to inference and learning of graph-based models in computer v...
The last couple of decades have seen a surge of interest and sophistication in using heuristics to s...
Knowing the probabilities of an object possible positions in a certain space and the constraints rel...
En informatique, la résolution de problèmes NP-difficiles en un temps raisonnable est d’une grande i...
National audienceRecently, deep neural networks have proven their ability to achieve excellent resul...
International audiencePlanning is the task of searching an action plan to achieve a goal. Classical ...
Co-localisées avec la Plate-Forme Intelligence Artificielle (PFIA 2019)International audienceTo pro...
L'optimisation des systèmes de classification est une tâche complexe qui requiert l'intervention d'u...
International audienceReinforcement learning algorithms have succeeded over the years in achieving i...
The manuscript is divided in two parts. The first consists in Chapters I to IV and offers a unified ...
National audienceDans cet article, nous présenterons les travaux prélimi-naires menés sur l'utilisat...
The impressive breakthroughs of the last two decades in the field of machine learning can be in larg...
L'apprentissage structuré est au fondement des méthodes modernes d'apprentissage automatique pour le...
La recherche efficace d'un chemin (quasi) optimal dans un graphe reste une tâche fondamentale en Inte...
The last couple of decades have seen a surge of interest and sophistication in using heuristics to s...
This thesis presents our contributions to inference and learning of graph-based models in computer v...
The last couple of decades have seen a surge of interest and sophistication in using heuristics to s...
Knowing the probabilities of an object possible positions in a certain space and the constraints rel...
En informatique, la résolution de problèmes NP-difficiles en un temps raisonnable est d’une grande i...
National audienceRecently, deep neural networks have proven their ability to achieve excellent resul...
International audiencePlanning is the task of searching an action plan to achieve a goal. Classical ...
Co-localisées avec la Plate-Forme Intelligence Artificielle (PFIA 2019)International audienceTo pro...
L'optimisation des systèmes de classification est une tâche complexe qui requiert l'intervention d'u...
International audienceReinforcement learning algorithms have succeeded over the years in achieving i...
The manuscript is divided in two parts. The first consists in Chapters I to IV and offers a unified ...
National audienceDans cet article, nous présenterons les travaux prélimi-naires menés sur l'utilisat...
The impressive breakthroughs of the last two decades in the field of machine learning can be in larg...
L'apprentissage structuré est au fondement des méthodes modernes d'apprentissage automatique pour le...