Grégoire Montavon
Grégoire Montavon works on explanation methods that show which inputs drive a neural network's predictions. He received a master's degree from École Polytechnique Fédérale de Lausanne in 2009 and completed a doctorate at Technische Universität Berlin in 2013. In 2015 he co-authored a PLOS ONE paper on layer-wise relevance propagation, and in 2017 he extended that method to recurrent networks such as LSTMs and GRUs. Pattern Recognition named him a recipient of its best paper award in 2020, and the 3rd World Conference on eXplainable Artificial Intelligence honored one of his papers in 2025. He is a professor at Charité – Universitätsmedizin Berlin and a research group lead at BIFOLD.
- Works
- 155
- Citations
- 17,112
- h-index
- 35
Publication numbers from OpenAlex, as of 15 October 2026.
Timeline
-
2009–2013
A machine learning doctorate
He receives a master's degree in communication systems from École Polytechnique Fédérale de Lausanne in 2009. He completes a doctorate in machine learning at Technische Universität Berlin in 2013.
-
2015
Pixel-wise explanations of classifiers
A 2015 PLOS ONE paper on pixel-wise explanations of classifier decisions by layer-wise relevance propagation lists him as a co-author.
-
2017
Explaining recurrent networks
In 2017, he co-authors a paper that extends layer-wise relevance propagation to recurrent networks such as LSTMs and GRUs.
-
2020
A Pattern Recognition best paper award
Pattern Recognition names him a recipient of its Best Paper Award in 2020.
-
2025
A best paper at XAI 2025
A paper of his wins the best paper award at the 3rd World Conference on eXplainable Artificial Intelligence in July 2025.
-
Since April 2025
A BIFOLD-Charité professorship
In April 2025, he takes up a BIFOLD-Charité professorship. He is a professor at Charité – Universitätsmedizin Berlin and a research group lead at BIFOLD.
Sources and credits
Photo: We have no published photo of Grégoire Montavon yet. Send us a free-licensed photo.