Laure Berti-Equille is a Research Director (DR1) at IRD (Institut de Recherche pour le Développement), based at the ESPACE-DEV joint research unit in Montpellier, France. Her work spans data quality, data preparation and cleaning, truth discovery and fact-checking, and trustworthy machine learning — including uncertainty quantification, anomaly detection, and multimodal deep learning — with applications to Earth observation and the United Nations Sustainable Development Goals.
She has published more than 200 papers in leading venues such as VLDB, SIGMOD, ICDE, KDD, ICLR, AISTATS and SIGIR, authored the open-access book AI for SDGs (EDP Sciences, 2025), and created open-source tools including Learn2Clean for reinforcement-learning-based data preparation. Before IRD she held research positions at MIT (visiting researcher, LIDS), the Qatar Computing Research Institute, AT&T Labs–Research, and the University of Rennes. She is a Senior Member of IEEE and ACM and a member of the ELLIS society.
Research themes
Data quality and cleaning · truth discovery and fact-checking · trustworthy and uncertainty-aware machine learning · anomaly detection in time series · multimodal deep learning · AI for the Sustainable Development Goals (Earth observation, biodiversity, poverty estimation).
Selected publications
- G. Bezirganyan, S. Sellami, L. Berti-Equille, S. Fournier. Classification with Uncertainty-Aware Multimodal Deep Learning: A Survey. ACM SIGKDD Explorations 28(1):41–62, 2026.
- L. Berti-Equille. Data Quality Profiling at Scale with Progressive Sampling: A Benchmark for Data-Centric AI Pipelines. Trans. on Large-Scale Data- and Knowledge-Centered Systems (TLDKS), Springer, 2026.
- S. Alnegheimish, D. Liu, C. Sala, L. Berti-Equille, K. Veeramachaneni. Sintel: An Overarching Ecosystem for End-to-End Time Series Anomaly Detection. ACM SIGMOD 2022.
- L. Berti-Equille. Learn2Clean: Optimizing the Sequence of Tasks for Web Data Preparation. The Web Conference (WWW) 2019.
- L. Berti-Equille, H. Harmouch, F. Naumann, N. Novelli, S. Thirumuruganathan. Discovery of Genuine Functional Dependencies from Relational Data with Missing Values. VLDB 2018.