CEA is also focused on KTT towards industry and society, and I have been involved over the last 25 years in Knowledge & Technology Transfer (KTT) activities at different levels:

• CEA and Industry: My main connections to industry are through European projects. I have been collaborating on R&D projects with SAGEM since 2007, first by supervising a PhD student paid by SAGEM, and then through two European projects (FP7 MC-IAPP CS-Orion (2010-2014), and FET DEDALE (2015-2018)). I have also worked with two other companies (IMEC and Planetek) on the H2020-COMPET-06-2014 PHySIS project. Since 2010, I have received  €5 million of European funding for R&D collaborations with industry. In addition, several of my softwares were packaged and distributed to private companies under a CEA licence.
• CEA Direction de la Recherche Technologique (DRT): the DRT unit does collaborative work with the industry in addition to research. I have supervised a PhD student, funded by the DRT, to work on mass spectroscopy and blind source separation, using the techniques we have developed in my group (Rapin et al 2013; 2014; 2016). This work had a wide range of applications, from material sciences to isotope tracing and dating, and from protein characterization to chemistry.
• CEA Life Science and Matter Science: in January 2016, the two fundamental research division (DSM - Science of Matter and DSC - Life Sciences) merged, creating a new division of Fundamental Research (DRF). The main idea was to cross-fertilize between both domains, and a CEA internal call for project was made. Our COSMIC project was one of few accepted. Its goal is to transfer techniques we developed for radio-interferometry to Magnetic Resonance Imaging (MRI). Our techniques will be applied NEUROSPIN data, a leading centre of neuro-imagery.
• Collaborations in Agriculture, Finance, Image and Video Processing, Statistics, etc. . . : I have collaborated with experts from a wide variety of fields. For example: economic and finan- cial measures used in agriculture (Morehart et al, 1999), financial data streams (Zheng et al, 1999), financial forecasting (Renaud et al, 2003; 2005) electricity forecasting (Benaouda, 2006), video processing (Tzagkarakis et al, 2012), laser range imaging (Tsagkatakis, 2015), optimization theory (Donoho et al, 2012), and clustering (Murtagh et al, 2000). These collaborations and papers used ideas and codes I had developed for astrophysics and are diverse examples of my KTT activities.
• CEA Astrophysics department: Knowledge transfer can occur even within two-sub fields that seem similar from the outside. The methods I developed for cosmology have many applications in astrophysics. Transferring code and knowledge to my immediate colleagues has always being a part of my activities at CEA. One striking example is the case of the exo-planet Kepler mission where light curves have missing data, making many observed stars data impossible to analyze. Thanks to my inpainting techniques, data that were thought to be lost could be reprocessed, analyzed and published (Pires et al 2015).
• Reproducible research: Reproducibility is at the heart of scientific methodology, and experiments must be reproducible for a result to be well established. Algorithms to extract information from data are more and more complex, and it has become challenging to reproduce others’ results. Since the statistical analysis can play a key role in any result, whether in cosmology or other fields, reproducible research is crucial for scientific advance. The concept of reproducible research consists in providing not only the data related to a given paper, but also codes that were used to analyse the data and the scripts that were used to process the data and create the figures. In this spirit, I put more than 25 packages freely available on the web (see here), and I encourage all my students and postdocs to follow this principle. My iSAP package was downloaded over 10k times. This shows that dedication to reproducible research supports KTT.