Postdoctoral Position in Deep Learning for Log-based Anomaly Detection and Domain Adaptation Université de Rouen in France

Postdoctoral position in deep learning, log-based anomaly detection, and domain adaptation within the ANR project SHARP (Machine Learning for Safe Vehicle Charging Points).

The research focuses on learning robust representations for highly heterogeneous log data, constructing reference distributions for normal behavior, and developing anomaly detection methods that remain effective under domain shift, drift, and non-IID conditions. The application domain is electric vehicle charging stations, with attention to operational anomalies, heterogeneous environments, and distributed settings.

The postdoc will be hosted at LITIS laboratory, Université de Rouen, and jointly supervised by Maxime Bérar (LITIS / Université de Rouen), Gilles Gasso (LITIS / INSA Rouen Normandie), and Mokhtar Z. Alaya (LMAC / Université de Technologie de Compiègne).

Eligibility highlights include a PhD in Applied Mathematics, Computer Science, Data Science, Machine Learning, or a related field; a strong publication record in machine learning or deep learning; hands-on experience with PyTorch and large-scale multi-GPU environments; and the ability to work independently and collaboratively.

The contract duration is 1 year, with a start date in September 2026. The monthly gross salary is stated as 2510€ to 2584€ depending on experience.

To apply, candidates should email a PDF package containing a cover letter, CV, publications or research papers, and contact details for at least two references to the three supervisors listed in the post.