- Paper 'One for all and all for one: Efficient Computation of Partial Wasserstein Distances on the Line' accepted at ICLR'25
- Paper 'Match-And-Deform: Time Series Domain Adaptation through Optimal Transport and Temporal Alignment' accepted at ECML'23
- Three new papers accepted on trajectory reconstruction from pen sensor data (IJDAR), inference of personality traits from wearable data (USENIX), and early classification for crop monitoring (ISPRS)
- Paper 'Time Series Alignment with Global Invariances' accepted in TMLR
- Blog posts on the differentiability of Dynamic Time Warping and Dynamic Time Warping
- Developed tslearn: a machine learning toolkit dedicated to time series data
Research Experience
- Conducts research at IRISA, focusing on machine learning and indexing for time series data
- Supervises multiple PhD students and post-docs
Background
Full professor at University of Rennes 2 since September 2022, following nine years as an assistant professor. Research focuses on machine learning and indexing for time series data, specifically for environmental time series.