Sociétés durables et résilientes

Le programme de recherche Sociétés Durables et Résilientes développe des solutions pilotées par l’intelligence artificielle pour anticiper et atténuer les bouleversements à l’échelle mondiale, qu’il s’agisse d’événements climatiques défavorables ou de crises géopolitiques. En analysant des données complexes issues de sources variées, telles que des images, de la parole et du texte, nos recherches permettent de prédire les risques et d’élaborer des stratégies d’atténuation fondées sur des preuves. Nous concevons des systèmes d’IA économes en énergie qui optimisent l’acquisition et la détection de données, permettant un déploiement sur des capteurs à faible coût et des dispositifs embarqués, tout en tenant compte à la fois des piliers économique, environnemental et sociétal de la durabilité. Notre travail lutte également contre la désinformation grâce à des modèles adaptatifs capables de synthétiser et de vérifier des éléments de preuve en temps réel.

Par une collaboration entre l’informatique, la physique, les sciences politiques et le droit, nous analysons les processus nuisibles et anticipons leur évolution et leur stabilité. Testées et validées dans des situations réelles, allant de l’évaluation des risques géopolitiques à l’optimisation des ressources en passant par la protection des infrastructures critiques, nos recherches proposent des solutions évolutives et impactantes pour des sociétés plus résilientes.

Domaines d’application

  • Détection & Atténuation de la désinformation
  • Technologies préservant la vie privée
  • Villes intelligentes, Efficacité énergétique, Durabilité & Confort urbain
  • Prédiction des risques & Systèmes d’information
  • Aide à la décision pour la gouvernance et les services publics

Domaines d'expertise

  • Bio-informatique & Informatique de la santé
  • Science des données & Informatique sociale
  • Interaction homme–machine
  • Imagerie & Vision par ordinateur
  • Apprentissage automatique
  • Traitement automatique du langage naturel
  • Robotique & Systèmes autonomes
  • Sécurité & Protection de la vie privée
  • Traitement du signal
  • Traitement de la parole & du signal audio

Plus  

Personnes

ABBET, Philip
(Senior Research and Development Engineer)
- website


AL AMINE, Zeina
(Research Intern)


ALMOMANI, Ahmad Qasim Mohammad
(Research Intern)


ALONSO DEL BARRIO, David
(Research and Development Engineer)


BEN MAHMOUD, Imen
(Research and Development Engineer)
- website


BHATTACHARJEE, Sushil (Kumar)
(Research Associate)
- website


BOGHETTI, Roberto
(Research Intern)
- website


BOLLENGIER, Alexis
(Research Intern)


BORNET, Olivier
(Head of Research and Development Team)
- website


BURDISSO, Sergio (Gastón)
(R&D / Research Assistant)


CANÉVET, Olivier
(Senior Research and Development Engineer)
- website


CAROFILIS VASCO, Roberto Andrés
(Postdoctoral Researcher)


CARRON, Daniel
(Senior Research and Development Engineer)
- website


CLIVAZ, Guillaume
(Senior Research and Development Engineer)
- website


COLBOIS, Laurent
(Postdoctoral Researcher)


DAYER, Yannick (Nicolas)
(Senior Research and Development Engineer)
- website


DROZ, William
(Senior Research and Development Engineer)
- website


ECABERT, Christophe
(Research Associate)


EL HAJAL, Karl
(PhD Student / Research Assistant)


GAIST, Samuel
(Senior Research and Development Engineer)
- website


GEISSBUHLER, David
(Research Associate)
- website


GEORGE, Anjith
(Research Associate)
- website


HERMANN, Enno
(Postdoctoral Researcher)
- website


HOVSEPYAN, Sevada
(Research Associate)


KÄMPF, Jérôme
(Senior Research Scientist)
- website


KAYAL, Salim
(Senior Research and Development Engineer)
- website


KHALIL, Driss
(Junior R&D / Research Assistant)
- website


KOMATY, Alain
(Research Associate)


KOTWAL, Ketan
(Research Associate)
- website


KRIVOKUĆA HAHN, Vedrana
(Research Associate)
- website


KULKARNI, Ajinkya (Vijay)
(Postdoctoral Researcher)


KULKARNI, Atharva (Abhijit)
(Research Intern)


KUMAR, Shashi
(PhD Student / Research Assistant)
- website


LI DONG, Virgílio
(Apprentice)


LUÉVANO GARCÍA, Luis Santiago
(Postdoctoral Researcher)
- website


MACEIRAS, Jérémy
(Senior Research and Development Engineer)
- website


MAGIMAI DOSS, Mathew
(Senior Research Scientist)
- website


MARCEL, Christine
(Senior Research and Development Engineer)
- website


MARCEL, Sébastien
(Senior Research Scientist with Academic Title, Interim management team)
- website


MAYORAZ, André
(Research and Development Engineer)
- website


MICHEL, Samuel
(Research and Development Engineer)
- website


MOHAMMADI, Amir
(Research and Development Engineer)


MOTLICEK, Petr
(Senior Research Scientist)
- website


NANCHEN, Alexandre
(Senior Research and Development Engineer)
- website


OTROSHI SHAHREZA, Hatef
(Postdoctoral Researcher)
- website


OUTUMURO BUENO, Alberto
(Research Intern)


ÖZTÜRK, Ünsal
(Postdoctoral Researcher)


POH, François
(Research Intern)
- website


POLAC, Magdalena
(Research Assistant)


PUROHIT, Tilak
(Research Intern)


RAHIMI NOSHANAGH, Parsa
(PhD Student / Research Assistant)


ULUCAN, Ibrahim
(Research Assistant)
- website


VERZAT, Colombine
(Senior Research and Development Engineer)
- website


VIDIT, Vidit
(Postdoctoral Researcher)


VILLATORO TELLO, Esaú
(Research Associate)
- website


WATAWANA, Hasindri (Sankalpana)
(PhD Student / Research Assistant)
- website


ZANGGER, Alicia
(Research and Development Engineer)
- website


 

Publications choisies

Integrating daylight with general and task lighting: A longitudinal in-the-wild study in individual and open space working areas, Chantal Basurto, Michael Papinutto, Moreno Colombo, Roberto Boghetti, Kornelius Reutter, Julien Nembrini and Jérôme Kämpf, in: Solar Energy Advances, 2, 2022.

This paper makes use of AI-based surrogate models to predict the indoor lighting conditions and control optimally the blinds and electric lighting to maintain visual comfort and achieve energy savings. More than 50% of electricity for lighting were saved without impacting significantly visual comfort over the course of our longitudinal experiment.

 

Comprehensive Vulnerability Evaluation of Face Recognition Systems to Template Inversion Attacks Via 3D Face Reconstruction, H. S. Otroshi and S. Marcel, IEEE TPAMI 2023.

In this work, we propose a new method (called GaFaR) to reconstruct 3D faces from facial templates using a pretrained geometry-aware face generation network, and train a mapping from facial templates to the intermediate latent space of the face generator network. We train our mapping with a semi-supervised approach using real and synthetic face images. For real face images, we use a generative adversarial network (GAN)-based framework to learn the distribution of generator intermediate latent space. For synthetic face images, we directly learn the mapping from facial templates to the generator intermediate latent code. We demonstrated the transferability of our attack with state-of-the-art methods across other face recognition systems. We also performed practical presentation attacks on face recognition systems using the digital screen replay and printed photographs, and evaluated the vulnerability of face recognition systems to different template inversion attacks.

 

Claim-Dissector: An Interpretable Fact-Checking System with Joint Re-ranking and Veracity Prediction, Martin Fajcik, Petr Motlicek and Pavel Smrz, in: Association for Computational Linguistics, Findings of the Association for Computational Linguistics: ACL 2023:10184–10205, 2023.

This paper describes new latent variable model for fact-checking and fact-analysis, which given a claim and a set of retrieved provenances allows learning jointly: (i) what are the relevant provenances to this claim (ii) what is the veracity of this claim. We propose to disentangle the per-provenance relevance probability and its contribution to the final veracity probability in an interpretable way - the final veracity probability is proportional to a linear ensemble of per-provenance relevance probabilities. This way, it can be clearly identified the relevance of which sources contributes to what extent towards the final probability. We show that our system achieves state-of-the-art results on FEVER dataset comparable to two-stage systems typically used in traditional fact-checking pipelines, while it often uses significantly less parameters and computation.

Projets choisis

Eguzki, 2020-2024, SFOE, KÄMPF: A simulation program for district heating networks based on artificial intelligence for the rapid and predictive resolution of complex looped networks

The project focuses on the pivotal role of district heating networks in harnessing lost heat for energy efficiency. It uses artificial intelligence to optimize network design, reduce costs, and minimize energy losses before significant investments are made.

 

TRESPASS, 2020-2024, H2020, MARCEL: Biometrics security and privacy preservation

The aim of this project is to combat rising security challenges with biometric technologies which are growing at a fast pace. More particularly, our researchers are investigating new types of security protection (e.g. presentation attack detection (PAD), morphing attack detection (MAD), deepfake detection (DD) or poisoning detection technologies) and privacy preservation (e.g. vulnerability assessment, template protection or computationally feasible encryption solutions).

 

CRiTERIA, 2021-2024, H2020, MOTLICEK: Comprehensive data-driven Risk and Threat Assessment Methods for the Early and Reliable Identification, Validation and Analysis of migration-related risks

The project aims to strengthen and expand existing risk analysis methods by introducing a novel, comprehensive but feasible and human-rights sensitive risk and vulnerability analysis framework for border agencies. The project started in 2021 and runs for three years. Idiap contributes to the project by developing innovative solutions automatically extracting relevant evidence from spoken and textual resources. Among technologies developed by Idiap are: (a) multilingual automatic speech recognition, (b) fact-checking system (i.e., system which can verify a claim formulated in natural language, whether it is true or not, by confirming against other factoid sources, and (c) reliability detector (i.e., a tool which can automatically evaluate the reliability of source (related to general OSINT data) as an unavoidable block for the fact-checking system.


 

Full list of related projects

Eguzkiand IVECT, 2020-2023, SFOE, Kämpf

Built environment sustainability

 

SOTERIA, 2022-2024, EU, Marcel

Face recognition anti-spoofing

 

GRAIL, 2022-2025, US IARPA, Marcel

Person recognition at a distance

 

TRESPASS, 2020-2024, EU, Marcel

Biometrics security and privacy preservation

 

CRiTERIA, 2021-2024, EU, Motlicek

Comprehensive data-driven risk and threat assessment methods for the early and reliable Identification, validation and analysis of migration-related risks

 

ROXANNE, 2019-2023, EU, Motlicek

Real-time network, text and speaker analytics for combating organized crime

 

TRACY, 2023-2025, EU, Motlicek

Big-data analytics from base-stations registrations and e-evidence system