Paris, France
+33 6 84 12 47 08
cyril.poulet@centraliens.net

Resume and papers

Experience

2016 – 2018 : Quicksign – R&D engineer / data scientist

Lead on R&D projects:

  • understanding of the company needs,
  • study of the state-of-the-art,
  • create POC: manage tooling, data annotation, trainings, design of larger solutions (including Deep Learning, Machine Learning, IA, Computer Vision, autooptimization), measure of performances and analysis
  • develop POC into automatable solution (configuration management, logging analysis, deployability and auto-build)
  • provide production-ready entity (Docker)

Subjects studied: Automated information extraction from images:

  • document localization in image
  • text reading
  • information tagging / text processing
  • information extraction / consolidation with knowledge of the trade.

Keywords: python, Tensorflow, IA, docker, image/text processing

2013 – 2015 : Evergig – R&D engineer

Study state-of-the-art IA research, choose technical solution, implement Proof-of-Concept, and integrate in company solution for various subjects:

  • Automated music mastering and equalization
  • Overlay detection on video
  • Noise reduction on multi-sources audio
  • Recommendation systems
  • Saliency and points of interest in video. Automated intelligent extraction of video thumbnails
  • Voice detection in audio segments
  • Video quality assessment
  • Concert set list extraction from Social media video titles

Keywords: C++, IA, video processing, optimization

2008 : Computational & Biological Learning Lab, CIMS, NYU – Intern

Development of a C++ GUI library for a convolutional neural network library (Libeblearn). Design, implementation and training of a neural net for face detection using these libraries. (Supervision : Prof. Yann LeCun)

 

Training

 

2010 – 2013 : University Paris VI (UPMC/LIP6) – PhD

PhD in Artificial Intelligence, on Multi-Agent coordination (with a particular interest on the patrolling problem with an open system setting)

2007 – 2008 : University Paris VI (UPMC) – Master 2

Signals, Systems, Images and Robotics, specialized in Robotics and Intelligent Systems (M2 SSIR/RSI)

2005 – 2008 : Ecole Centrale Paris – Engineering degree

Prestigious university-level French multidisciplinary engineering school. Specialisation in Research, conception and development in the field of Advanced Electronic Systems (SA/CDR)

 

Published Papers

 

[8] C. Poulet, V. Corruble and A. El Fallah Seghrouchni, “Working as a team: using social criteria in the timed patrolling problem” , in 24th IEEE International Conference on Tools with Artificial Intelligence (ICTAI), , IEEE, Athene, 2012. ictai2012

[7] C. Poulet, V. Corruble and A. El Fallah Seghrouchni, “Travailler en équipe : le choix social appliqué au problème de la patrouille multi-agents” , in Actes des journées francophones sur les systèmes multi-agents (JFSMA ’12), Honfleur, 2012. jfsma2012

[6] C. Poulet, V. Corruble and A. El Fallah Seghrouchni, “Auction-based strategies for the open-system patrolling task” , in Proc. PRIMA – 15th International Conference on Principles and Practice of Multi-Agent Systems, Springer, Kuching, Malaysia, 2012. prima2012

[5] C. Poulet, V. Corruble, A. El Fallah Seghrouchni and G. Ramalho, “The Open System Setting in Timed Multiagent Patrolling” , in Proc. of Web Intelligence – Intelligent Agent Technologies (WI-IAT’11), IEEE, Lyon, 2011. IAT2011

[4] C. Poulet, “Etude du Problème de la Patrouille Multi-Agents en Système Ouvert” , in Proc. of Rencontre des Jeunes Chercheurs en Intelligence Artificielle (RJCIA’11), Grenoble, 2011. RJCIA2011

[3] C. Farabet, C. Poulet and Y. LeCun, “An FPGA-Based Stream Processor for Embedded Real-Time Vision with Convolutional Networks”, in Proc. of the Fifth IEEE Workshop on Embedded Computer Vision (ECV’09 @ ICCV’09), IEEE, Kyoto, 2009. ecvw09

[2] C. Farabet, C. Poulet, J. Y. Han, and Y. LeCun, “CNP: An FPGA-based Processor for Convolutional Networks”, in Proc. International Conference on Field Programmable Logic and Applications (FPL’09), IEEE, Prague, 2009. fpl09

[1] C. Farabet, C. Poulet, J. Y. Han, and Y. LeCun, “An FPGA-based Processor for Convolutional Networks”, in Snowbird Learning Workshop, Clearwater FL, 2009.