The LIAD team
Geoffrey DANIEL
Engineer-researcherGeoffrey is a research engineer specialising in machine learning and its application to scientific data processing.
He obtained his doctorate in 2020 from the Université de Paris-Cité for his thesis work at the CEA on the subject: "Development and optimisation of a miniature Compton camera with coded mask: method for analysing a radiative environment by spectro-identification and 3D localisation of gamma-ray sources". The core of the work consisted of applying machine learning methods to the analysis of gamma camera data for both spectroscopy and imaging.
Following his PhD, Geoffrey joined the Artificial Intelligence and Data Science Laboratory team. His activities cover both theoretical topics, such as uncertainty quantification for neural network predictions or the robustness of machine learning models, and applications. He works in collaboration with other teams at the CEA on the use of AI methods for scientific data analysis and is involved in various projects, such as the European MatCHMaker project dedicated to the development of sustainable materials.
He also coordinates the ALLEGRIA network, which proposes scientific exchanges and seminars on AI applications at the CEA, and he teaches courses in probability and machine learning at the INSTN.
Publications list (non-exhaustive)
- G. Daniel et al. Deep learning reconstruction with uncertainty estimation for γ photon interaction in fast scintillator detectors, Engineering Applications of Artificial Intelligence, Volume 131 (2024)
- F.M.F. de Oliveira, G. Daniel, O. Limousin Artificial gamma ray spectra simulation using Generative Adversarial Networks (GANs) and Supervised Generative Networks (SGNs), Nuclear Instruments and Methods in Physics Research Section A, Volume 1047 (2023)
- R. Le Breton, O. Limousin, G. Daniel, et al. The Spid-X gamma camera: A miniature gamma ray integral field spectrometer for nuclear industry applications, Nuclear Instruments and Methods in Physics Research Section A, Volume 1047 (2023)
- O. Laurent, A. Lafage, E. Tartaglione, G. Daniel, J.-M. Martinez, A. Bursuc, G. Franchi Packed-Ensembles for Efficient Uncertainty Estimation, Published as a conference paper at ICLR 2023 (2023)
- Z. Chaouai, G. Daniel, J.-M. Martinez, O. Limousin, A. Benoit-Lévy Application of adversarial learning for identification of radionuclides in gamma-ray spectra, Nuclear Instruments and Methods in Physics Research Section A, Volume 1033 (2022)
- G. Daniel, Y. Gutierrez, O. Limousin Application of Deep Learning algorithm to Compton imaging of radioactive point sources with a single planar CdTe pixelated detector, Nuclear Engineering and Technology, Volume 54, Issue 5, pp. 1747-1753 (2022)
- G. Daniel, O. Limousin Extended sources reconstructions by means of coded mask aperture systems and deep learning algorithm, Nuclear Instruments and Methods in Physics Research Section A, Volume 1012 (2021)
- G. Daniel, F. Ceraudo, O. Limousin, D. Maier, A. Meuris Automatic and Real-time Identification of Radionuclides in Gamma-Ray Spectra: A new method based on Convolutional Neural Network trained with synthetic data set, IEEE Transactions on Nuclear Science, Vol. 67, No. 4, pp. 644-653 (2020)
- G. Daniel, O. Limousin, D. Maier, A. Meuris, F. Carrel Compton imaging reconstruction methods: a comparative performance study of direct back-projection, SOE, a new Bayesian algorithm and a new Compton inversion method applied to real data with Caliste, EPJ Web Conf. ANIMMA 2019, Volume 225 (2020)
- D. Maier, O. Limousin, G. Daniel Energy calibration via correlation using an adaptive mesh refinement, Web Conf. ANIMMA 2019, Volume 225 (2020)
- D. Maier, G. Daniel et al. Second generation of portable gamma camera based on Caliste CdTe hybrid technology, NIM-A, Volume 912, pp. 338-342 (2018)