Lerasle, Matthieu Szabó, Zoltán Lecué, Guillaume Massiot, Gaspar Moulines, Eric

Mean embeddings provide an extremely flexible and powerful tool in machine learning and statistics to represent probability distributions and define a semi-metric (MMD, maximum mean discrepancy ; also called N-distance or energy distance), with numerous successful applications. The representation is constructed as the expectation of the feature map...

CHOLET, Stephane Paugam-Moisy, Helene

International audience

Cambuzat, Remi Elisei, Frédéric Bailly, Gérard Simonin, Olivier Spalanzani, Anne

This paper presents a new teleoperation system – called stereo gaze-contingent steering (SGCS) – able to seamlessly control the vergence, yaw and pitch of the eyes of a humanoid robot – here an iCub robot – from the actual gaze direction of a remote pilot. The video stream captured by the cameras embedded in the mobile eyes of the iCub are fed into...

PIERROT, David Harbi, Nouria Darmont, Jérôme

Les conséquences d'une intrusion dans un système d'information peuvent s'avéver problématiques pour l'existence d'une entreprise ou d'une organisation. Les impacts sont synonymes d'une perte financière, d'image de marque et de sérieux. La détection d'une intrusion n'est pas une finalité en soit, la ré-duction du delta détection-réaction est devenue...

Bigand, André Constantin, Joseph Renaud, Christophe Dehos, Julien

MINOT, Maël

The objective of this thesis is, from a general standpoint, to design and evaluate decomposition methods for solving constrained optimisation problems. Two optimisation problems in particular are considered: the maximum common induced subgraph problem, in which the largest common part between two graphs is to be found, and the sum colouring problem...

Calandriello, Daniele

The main advantage of non-parametric models is that the accuracy of the model (degreesof freedom) adapts to the number of samples. The main drawback is the so-called "curseof kernelization": to learn the model we must first compute a similarity matrix among allsamples, which requires quadratic space and time and is unfeasible for large datasets.Non...

Yoshikawa, Nobuyuki Belkhir, Nacim Suzuki, Sinji

This paper empirically investigate the design of a fault detection mechanism based on Long Short Term Memory (LSTM) neural network. Given an equation based model that approximate the behavior of aircraft ailerons, the fault detector aims at predicting the state of aircraft: the normal state for which no failure are observed, or four different failu...

Jitkrittum, Wittawat Xu, Wenkai Szabó, Zoltán Fukumizu, Kenji Gretton, Arthur

We propose a novel adaptive test of goodness-of-fit, with computational cost linear in the number of samples. We learn the test features that best indicate the differences between observed samples and a reference model, by minimizing the false negative rate. These features are constructed via Stein’s method, meaning that it is not necessary to comp...

Pageaud, Simon Deslandres, Véronique Lehoux-Lebacque, Vassilissa Hassas, Salima

Designing a public urban policy is a demanding process which requires both time and money with no warranty of its efficiency. It involves knowledge about the purpose of urban design, behaviors of users and needs in terms of mobility. We believe that in the near future, decision makers will have to react and more frequently adapt public policies, ba...