We introduce CST, an algorithm for constructing skill trees from demonstration trajectories in continuous reinforcement learning domains. CST uses a changepoint detection method to segment each trajectory into a skill chain by detecting a change of appropriate abstraction, or that a segment is too complex to model as a single skill. The skill chains from each trajectory are then merged to form ...

L’écoute attentive est plus que «ne rien faire»; il s’agit d’une offre consistant à se tourner vers les personnes concernées avec la grande attention et attendre si comment elles souhaitent poursuivre le dialogue.

We describe CST, an online algorithm for constructing skill trees from demonstration trajectories. CST segments a demonstration trajectory into a chain of component skills, where each skill has a goal and is assigned a suitable abstraction from an abstraction library. These properties permit skills to be improved efficiently using a policy learning algorithm. Chains from multiple demonstration ...

We describe recent work on CST, an online algorithm for constructing skill trees from demonstration trajectories. CST segments a demonstration trajectory into a chain of component skills, where each skill has a goal and is assigned a suitable abstraction from an abstraction library. These properties permit skills to be improved efficiently using a policy learning algorithm. Chains from multiple...

This work draws on previous works regarding spectral learning algorithm for structured data (see [5], [9], [2], [1], [8]). We present an extension of the Hidden Markov Models, called Graphical Weighted Models (GWM), whose purpose is to model distributions over labeled graphs. We describe the spectral algorithm for GWM, which generalizes the previous spectral algorithms for sequences and trees. ...

Résumé. Le choix du taux d’élagage est crucial dans le but d’acquérir une terminologie de qualité à partir de corpus de spécialité. Cet article présente une étude expérimentale consistant à déterminer le taux d’élagage le plus adapté. Plusieurs mesures d’évaluation peuvent être utilisées pour déterminer ce taux tels que la précision, le rappel et le Fscore. Cette étude s’appuie sur une autre me...

We develop a method for counting words subject to various restrictions by finding a combinatorial interpretation for a product of formal sums of Laguerre polynomials. We use this method to find the generating function for k-ary words avoiding any vincular pattern that has only ones. We also give generating functions for k-ary words cyclically avoiding vincular patterns with only ones whose runs...

The solid-state electrical conductivities of a number of ferredoxin model compounds are reported. For one of these compounds, (KFeS(2))(n), an electron transfer rate for a 25 A unit is shown to be at least 1 x 10(8) electrons sec(-1). The rate becomes proportionally larger for smaller molecular units. This rapid rate is consistant with a short pipe model for electron transport between two react...

Nous présentons une nouvelle approche robuste à la surestimation de la longueur du canal pour l'identi cation aveugle des canaux multiples RIF (à réponse impulsionnelle nie). L'approche prédiction linéaire présentée dans [1] est une méthode d'estimation robuste puisqu'elle permet de trouver un estimateur consistant du canal même si son ordre a été surestimé. Dans cette contribution, nous porton...