Scholar citations


  • L. Nardi, F. Badran, P. Fortin, J. Brajard and S. Thiria. Automatic Generation of Parallel and Coherent Code Using the YAO Variational Data Assimilation Framework. In review, 2016. [PDF] [bibtex]
  • Bercea, G.-T., McRae, A. T. T., Ham, D. A., Mitchell, L., Rathgeber, F., Nardi, L., Luporini, F., and Kelly, P. H. J.: A numbering algorithm for finite elements on extruded meshes which avoids the unstructured mesh penalty, Geosci. Model Dev. Discuss., September 2016. [PDF


  • L. Nardi, B. Bodin, S. Saeedi, E. Vespa, A. J. Davison and P. H. J. Kelly. Algorithmic Performance-Accuracy Trade-off in 3D Vision Applications Using HyperMapper. In IEEE Intl. Workshop on Automatic Performance Tuning (iWAPT), hosted at 31st IEEE Intl. Parallel & Distributed Processing Symposium (IPDPS), May 2017. [IEEE] [arXiv][PPT]
  • S. Saeedi, L. Nardi, E. Johns, B. Bodin, P. H. J. Kelly and A. J. Davison. Application-oriented Design Space Exploration for SLAM Algorithms. In IEEE Intl. Conf. on Robotics and Automation (ICRA 2017), Singapore, May 2017. [IEEE][PDF][PPTSpotlight][PPTInteractive][video].
  • B. Bodin, L. Nardi, Zia Zeeshan, H. Wagstaff, G. S. Shenoy, M. Emani, J. Mawer, C. Kotselidis, A. Nisbet, M. Lujan, B. Franke, Paul H. J. Kelly, M. O’Boyle. Integrating algorithmic parameters into benchmarking and  design space exploration in 3D scene understanding. In Intl. Conf. on Parallel Architectures and Compilation Techniques (PACT 2016), Haifa, September 2016. [ACM][PDF][PPT]
  • B. Bodin, L. Nardi, P. H. J. Kelly and M. O’Boyle. Diplomat: mapping of multi-kernel applications using a static dataflow abstraction. In Modelling, Analysis and Simulation of Computer and Telecommunication Systems (MASCOTS 2016), London, , September 2016. [IEEE][PDF]

  • M. Z. Zia, L. Nardi, A. Jack, E. Vespa, B. Bodin, P. H. J. Kelly and A. J. Davison. Comparative Design Space Exploration of Dense and Semi-Dense SLAM. In IEEE Intl. Conf. on Robotics and Automation (ICRA 2016), Stockholm, Sweden, May 2016. [IEEE][arXiv][PPTSpotlight][PPTInteractive]

  • A. Dietrich, S. Zug, L. Nardi and J. Kaiser. Reasoning in complex environments with the SelectScript declarative language. Intl. Workshop on Domain-Specific Languages and models for ROBotic systems (DSLRob), Hamburg, Germany, October 2015. [PDF][bibtex]

  • L. Nardi, B. Bodin, M. Z. Zia, J. Mawer, A. Nisbet, P. H. J. Kelly, A. J. Davison, M. Luján, M. F. P. O’Boyle, G. Riley, N. Topham, and S. Furber. Introducing SLAMBench, a performance and accuracy benchmarking methodology for SLAM. In IEEE Intl. Conf. on Robotics and Automation (ICRA 2015), Seattle, Washington USA, May 2015. [IEEE][arXiv][PPTSpotlight][PPTInteractive][bibtex]
  • L. Nardi, F. Badran, P. Fortin and S. Thiria. YAO: a generator of parallel code for variational data assimilation applications. Intl. Conf. on High Performance Computing and Communications (HPCC), 2012. [IEEE] [PDF] [PPT] [bibtex]
  • L. Nardi, C. Sorror, F. Badran and S. Thiria. YAO: A Software for Variational Data Assimilation Using Numerical Models. Intl. Conf. on Computational Science and its Applications (ICCSA), Seul, South Korea, June 2009. [Springer] [PDF] [PPT] [bibtex]


  • L. Nardi. Formalisation et automatisation de YAO, générateur de code pour l’assimilation variationnelle de données. Ph.D.
    thesis, CNAM and UPMC, Paris, France, March 2011. [PDF] [PPT] [bibtex]

Seminars and Workshops

  • Multi-objective Design Exploration and 
Its Application to Algorithmic Optimisation in Real-time Robot Vision, seminar Stanford, Palo Alto, January 2017. [PPT]
  • A Multi-objective Adaptivity Methodology Vertically Integrating Algorithmic Parameters and Design Space Exploration, seminar Argonne National Lab, Chicago, January 2017. [PPT]
  • Software Self-configurability in GPU-accelerated Robot Vision. Invited speaker at NVIDIA GTC-EU 2016 Amsterdam, September 28, 2016. [PPT][video_TBD]
  • Towards performance and accuracy adaptivity in GPU-accelerated 3D Robot Vision. Invited talk Politecnico di Milano, Italy, September 21, 2016. [PPT]
  • Holistic benchmarking in 3D Robot Vision: the SLAMBench open source framework. Invited talk HiPEAC CSW, Porto, April 21, 2016. [PPT]
  • Searching for the optimal algorithmic and implementation configuration in 3D robot vision. Invited talk HARSH workshop in conjunction with HPCA/CGO/PPoPP, Barcelona, March 12, 2016. [PPT]
  • Auto-tuning for algorithmic and implementation design choices in GPU-accelerated 3D robot vision. Invited talk Many-core seminar series, Oxford, UK, February 24, 2016. [PPT]
  • Vertically-integrated exploration of algorithmic and implementation design spaces in 3D scene understanding. Invited talk IFIP 2.11, London, November 11, 2015. [PPT]
  • Multi-space multi-objective 
 design-space exploration for 3D scene understanding using active learning. Imperial College Department of Computing RA Symposium, London, June 11, 2015. [PPT]
  • Get the Abstraction Right and Eat the Free Lunch: 
A Performance Perspective. Invited talk IBM TJ Watson, New York, May 22, 2015. [PPT]
  • Computing fast, computing smart. Invited talk ECMWF Reading, UK, May 11, 2015. [PPT]
  • A performance, energy and accuracy aware benchmarking methodology for robot vision. Invited talk Google Mountain View, California, March 24, 2015. [PPT]
  • A performance, energy and accuracy aware benchmarking methodology for robot vision. Invited talk AMD Sunnyvale, California, March 23, 2015. [PPT]
  • A performance, energy and accuracy aware benchmarking methodology for robot vision. Invited speaker at NVIDIA GTC 2015 San Jose, California, March 17, 2015. [PPT][video]
  • SLAMBench, a performance and accuracy benchmarking methodology for SLAM. Invited talk La Sapienza Rome, Italy, December 2014. [PPT]
  • YAO: toward an efficient data assimilation environment (extended). LIP6 Paris, France, March 2010. [PPT]
  • YAO: toward an efficient data assimilation environment. Ninth European Workshop on Automatic Differentiation (INRIA) Sophia Antipolis, France, November 2009. [PPT]
  • YAO, a software for numerical models and variational data assimilation. CNAM, Paris, France, June 5, 2008. [PPT]



  • L. Nardi. Modeling attack processes on the Internet, based on honeypot collected data – Development of a high-iteration honeypot data processing system. Master thesis, LAAS-CNRS and La Sapienza, Toulouse, France, April 2007. [PDF]
  • L. Nardi. Doppio visualizzatore di temperatura: 2µS1T2. Bachelor thesis, at Selcon Elettronica srl. and La Sapienza, Sezze scalo (LT), Italy, October 2002. [PDF]