Nikita Lagrange

Nikita Lagrange

PhD Student

About Me

📢 I am on the job market and planning to defend my PhD thesis in Fall 2025 — feel free to reach out !

I am Nikita Lagrange, a PhD student at Institut Curie in Paris under the supervision of Dr. Hervé Isambert and the co-supervision of Dr. Barbara Bravi.

My thesis focuses on the development of a new search-and-score algorithm for causal discovery in the presence of latent variables using information theory.

Download CV
Interests
  • Causal Discovery
  • Machine Learning
  • Probabilistic Methods
  • Information Theory
  • Bioinformatics
Education
  • PhD in Computer Science

    Sorbonne University

  • M.Sc. in Bioinformatics & Modelling

    Sorbonne University

  • B.Sc. in Life Sciences

    Sorbonne University

📚 Research

I am a PhD student at Institut Curie in Paris. My thesis focuses on causal discovery. More specifically, I am working on extending the MIIC algorithm, developed within the Isambert Lab to a search-and-score mode.

The goal is to improve the efficiency and robustness of causal discovery algorithms, particularly when applied to complex and non-linear biomedical data.

Beyond that, I am interested in developing machine learning algorithms that are more grounded in human reasoning and are interpretable.

📝 Publications
(2025). An Efficient Search-and-Score Algorithm for Ancestral Graphs using Multivariate Information Scores for Complex Non-linear and Categorical Data. Forty-second International Conference on Machine Learning (ICML 2025).
(2025). Hybridation et pédagogie par projet : retour d’expérience. Colloque “Questions de Pédagogies dans l’Enseignement Supérieur” (QPES2025).
(2025). Preserving information while respecting privacy through an information theoretic framework for synthetic health data generation. npj Digital Medicine.
(2025). CausalXtract, a flexible pipeline to extract causal effects from live-cell time-lapse imaging data. eLife.
Talks
Academic Services

Representative of doctoral students on the EDITE doctoral school board
Participation in the board and involvement in key decision-making processes.

Reviewer for the NeurIPS BDU Workshop 2024

📰 News
📅 May 1, 2025

Our method for search-and-score learning of ancestral graphs has been accepted at ICML 2025 in Vancouver 🇨🇦 ! Looking forward to presenting this work 🎉 !

📅 March 11, 2025

I presented my PhD thesis in 3 minutes during the EDITE 2025 Doctoral Day at the Centre Pompidou in Paris 🇫🇷 . An exciting challenge in science communication 📣 !

📅 February 21, 2025

Our contribution on a feedback from a new hybrid learning approach has been accepted at QPES2025 in Brest 🇫🇷 🎉 ! This work results from a data analysis consulting mission at CAPSULE, Sorbonne University 🎓

Contact

📧 nikita.lagrange@curie.fr or nikita.lagrange@tutanota.com

📍Institut Curie, Research Center, CNRS-UMR168
Pavillon Curie, 1st floor, Office 01.09.b
11, rue Pierre et Marie Curie
75005 Paris