--------

Modèles Informatiques du Langage et de la Cognition - MILC

Artificial Intelligence Group
Computer Science Department

 Français Research Papers Home


Alain GRUMBACH

Professor, Manager of the MILC Team, in charge of the ICC cursus

Biography

Received his State Engineering Degree from the Ecole Nationale Supérieure de l'Aéronautique et de l'Espace, Paris in 1970, and his Doctorat d'Etat in 1987. Held an Engineering position at the Marcoussis Research Center of the Compagnie Générale d'Electricité , then became Professor at the Ecole Supérieure d'Electricité. Currently Professor at ENST Paris within the Computer Science Department, his interests lie in : Artificial Intelligence, Cognitive Science, Connectionism.

Research

Our research topic are related to two main themes:
- A connectionist approach to time series prediction
- Tele-Operation in virtual environements: Helping the operator with cognitive models of his intentions

A connectionnist Approach to Time Series Forecasting

Nowadays, classical connectionnist techniques are very well-known, particularly in the field of MultiLayers Perceptrons (MLP). Nonetheless, time series predictions using MLP are quite difficult and results are not as good as they were supposed to be. In fact, time series do not accomodate very well with neural networks because of : The aim of this Ph.D. is to design a methodology for this kind of neural networks based forecasting, to find architectures which are suitable for some kind of prediction (FIR, TDNN, RST,... , hybrid ones), or design new ones, and finally to design tests which may be used to compare our results with statistical models.

Another part of our work is to find a new architecture dedicated to time dependant series, that is to say a neural network architecture where time will not be another spatial dimension. Opposed to TDNN, we want time to be a built-in dimension. A first attempt in this direction is the delta-NARMA neural network, which can be seen as a connectionist extension of ARIMA and ARARMA statistical models.

Another topic of interest for us is the fusion between symbolic and numerical processing in the particular case of time series prediction. We study both the use of classical statistical methods (Fischer statistic) and connectionist one to estimate relevance and redundancy. We also study from a more general point of view how to handle symbolic data with a classical connectionist network.

To implement delta-NARMA neural networks and to make experiments about the fusion of symbolic and numerical processing, we have developped a simulator: dnns.

Tele-Operation in virtual environements:
Helping the operator with cognitive models of his intentions

This project is in the field of Virtual Reality for robotics. It deals with the communication between a user and a virtual environment, in tele-operation context. We study a possibility of communication at the user intention level, in order to make operations easier for him. When a tricky operation (such as object prehension) is going to be done by the operator, the system detects the operator's intention through movements, distances, or other relations. After confirmation, it starts a process aimed at performing this operation automatically. When the operation is finished, the user get back the control of the robot. This project is mainly situated in the field of cognitive modeling : perceptions, actions, intentions. On the technical level, this project leads us to design, for each functionnality of the system, the most appropriated AI techniques (symbolic or connectionist), and a collaboration between them.

Papers

Neural Networks

Cognitive Science


Last changes May 30 1997
Alain Grumbach (grumbach@inf.enst.fr)