The project ODD is an ANR projet under the “Single-team research project” instrument (PRME). Although such an instrument is for local projects, within a given research group of the lab, we have designed the ODD project in order to go beyond the local dimension. <\/p>\n\n\n\n
Besides the research results<\/a> obtained by the team, the ODD project have structured the research activity in anomaly detection in the LITIS’s Machine Learning group. This organization of our team, and the consolidation of our ODD skills, have enabled us to become driving forces in setting up other highly-related research projects, including industrial and international openings.<\/p>\n\n\n\n The main projects are as follows.<\/p>\n\n\n\n The SHARP project addresses anomaly and change point detection for cyber physical security, with a focus on charging points for electric vehicles.<\/p>\n\n\n\n The project SHARP is funded by the ANR program “Th\u00e9matiques Sp\u00e9cifiques en Intelligence Artificielle” (TSIA). Its consortium is:<\/p>\n\n\n\n Several members of the ODD project are involved in SHARP: G. Gasso, P. Honeine, M. Berar and F. Pacheco, as well as M. Alaya (UTC). They will bring the ODD expertise to a novel application field, which is cybersecurity. The industrial partner of the project is Citeos Solutions Digitales.<\/p>\n\n\n\n The grant for the LITIS Lab includes 1 PhD and 1 Post-doc researcher.<\/p>\n\n\n\n The Share2Detect project addresses anomaly detection and domain adaptation in the context of federated learning, namely collaborative learning.<\/p>\n\n\n\n It is funded by the ANR “Young researchers” instrument. The project’s principal investigator is F. Pacheco, with a team including members of the ODD project (M. Berar and P. Honeine). The main objective is to extend the ODD expertise to the setting of federated learning.<\/p>\n\n\n\n The grant for the LITIS Lab includes a PhD and an 18-month post-doc researcher.<\/p>\n\n\n\n We extended the projet to international collaborations, as expected in the proposal of the ODD project. For this purpose, two projects were implemented, bringing together researchers from several universities to work on the objectives of the ODD project, which strengthened and consolidated its international outcomes and impacts. <\/p>\n\n\n\n The DD-AnDet project seeks to address anomaly detection for time series in order to analyze the vibration signals, which is an important application addressed in ODD. The involved universities are:<\/p>\n\n\n\n All the French contributors of the DD-AnDet project are members of the ODD project: F. Pacheco (project coordinator), P. Honeine and R. Mussard.<\/p>\n\n\n\n The DEEDS project addresses out-of-distribution detection in streaming data, extending the work of ODD to the ensemble learning formalism. The two involved universities are:<\/p>\n\n\n\n All the contributors of the DEEDS project are members of the ODD project.<\/p>\n","protected":false},"excerpt":{"rendered":" The project ODD is an ANR projet under the “Single-team research project” instrument (PRME). Although such an instrument is for local projects, within a given research group of the lab, we have designed the ODD project in order to go beyond the local dimension. Besides the research results obtained by the team, the ODD project … Continue reading OUTCOMES & IMPACTS<\/span> SHARP: Machine Learning for Safe Vehicle Charging Points (ANR, 2024-2028)<\/h2>\n\n\n\n
\n
Share2Detect: Sharing is detecting in Federated Learning (ANR, 2026-2030)<\/h2>\n\n\n\n
International Projects<\/h2>\n\n\n\n
DD-AnDet: Data-driven Anomaly Detectors for Time Series Data and Big Data (STIC-AmSud, 2024-2026):<\/h5>\n\n\n\n
\n
DEEDS: Deep Ensembles for Evolving Data Streams (CAPES COFECUB, 2025-2029)<\/h5>\n\n\n\n
\n