François Lanusse

Flanusse

I am currently a postdoc at the McWilliams Center for Cosmology at Carnegie Mellon University.

I have been part of the CosmoStat team at CEA Saclay from 2012 to 2015, where I have done my PhD under the supervision of Jean-Luc Starck.

Contact Information:
E-mail: flanusse@andrew.cmu.edu

Research Interests

My PhD subject is: “Sparse Reconstruction of the Dark Matter Mass Map”. Available here.

My research interests encompass both cosmology and astrostatistics with a focus on weak gravitational lensing and sparsity based methodologies. I am involved in the following projects:

  • Sparse reconstruction of 3D dark matter maps from weak lensing. We have developed a compressed sensing approach to the problem of 3D dark matter map reconstruction from weak lensing (GLIMPSE). Such maps can potentially lead the way to higher order statistics from weak lensing surveys to improve the constraints on cosmological parameters.
  • Primordial Power Spectrum Reconstruction. Based on the framework of sparse regularisation of linear inverse problems, we have developed a new approach to reconstruct the Primordial Power Spectrum from the anglar power spectrum of the CMB. Our new algorithm (PRISM) is capable of recovering the broad shape of the spectrum as well as isolated features.
  • Spherical 3D analysis of the galaxy clustering. We have developed 3D Spherical Wavelets well suited to the analysis of data provided in spherical coordinates. These wavelets are based on the Spherical Fourier-Bessel transform and can be applied to the study of large galaxy surveys (MRS3D). We have also investigated the impact of performing a full 3D Fourier-Bessel analysis of spectroscopic galaxy surveys on cosmological parameter constraints (CosmicPy).

Teaching Activities

  • Tutorials at the IAU Symposium 306: Statistical Challenges in 21st Century Cosmology, Lisbon, Portugal, 25-29 May 2014. 3 hours presentation and exercises on Sparse regularization of inverse problems
  • Tutorials at the 2nd Workshop on Scientific Computing in Astronomy, Sao Paulo, Brazil, 3-6 June 2014. 4 hours presentations and exercises on Wavelets, Sparsity, Denoising, Deconvolution and Morphological diversity.
  • Practical Labs at the 2nd International Summer School on INtelligent Signal Processing for FrontIEr Research and Industry (Infieri), Paris, France, 14-25 July 2014. 10 afternoon sessions of 3 hours on Introduction to sparsity

Education

I graduated from Supélec in 2011 with an Electrical Engineering degree and an MSc in Applied Mathematics. I obtained a second MSc in Fundamental Physics with a specialization in Cosmology at Paris 11 University in 2012 (Master Noyaux Particules Astroparticules et Cosmologie).

Publications

Book Chapter:

  • Lanusse, F., Starck, J.-L., Woiselle, A. and Fadili, M.J, 3D Sparse Representations, Advances in Imaging and Electron Physics, Advances in Imaging and Electron Physics , Academic Press: Elsevier Inc, pp. 99?204, 2014.

Refereed Articles:

  • Möller, A., Ruhlmann-Kleider, V., Lanusse, F., Neveu, J., Palanque-Delabrouille, N., 2015, SNIa detection in the SNLS photometric analysis using Morphological Component Analysis. Accepted in JCAP, arxiv:1501.02110
  • Leonard, A., Lanusse, F., Starck, J.-L.,2015, Weak lensing reconstructions in 2D & 3D: implications for clusters studies. MNRAS,449,1146
  • Lanusse, F., Rassat, A., Starck, J.-L., 2015, 3D Galaxy Clustering with Future Wide-Field Surveys: Advantages of a Spherical Fourier-Bessel analysis. Accepted in A&A, arxiv:1406.5989
  • Lanusse, F., Paykari, P., Starck, J.-L., Sureau, S., Bobin, J., Rassat, A., 2014, PRISM: Recovery of the primordial spectrum from Planck data. A&A, 571, id.L1
  • Paykari, P., Lanusse, F., Starck, J.-L., Sureau, S., Bobin, J., 2014, PRISM: Sparse Recovery of the Primordial Power Spectrum. A&A,566,A77
  • Leonard, A., Lanusse, F., Starck, J.-L.,2014, GLIMPSE: Accurate 3D weak lensing reconstructions using sparsity. MNRAS,440,1281
  • Lanusse, F., Rassat, A., Starck, J.-L.,2012, Spherical 3D Isotropic Wavelets. A&A,540,A92

Conference Proceedings:

  • Lanusse, F., Leonard, A., Starck, J.-L., 2014. Density reconstruction from 3D lensing: Application to galaxy clusters. Statistical Challenges in 21st Century Cosmology, Lisbon, Protugal, May 2014.
  • Lanusse, F., Leonard, A., Starck, J.-L., 2014. High resolution density reconstruction from 3D lensing. Application to galaxy clusters Rencontres de Moriond, Cosmology 2014, La Thuile, Italy, March 2014.
  • Lanusse, F., Starck, J.-L., 2014. 3D sparse representations on the sphere and applications in astronomy. Wavelet & Sparsity, SPIE Optics and Photonics, San Diego, USA, August 2014.
  • Lanusse, F., Leonard, A., Starck, J.-L., 2014. Imaging Dark Matter using Sparsity. Wavelet & Sparsity, SPIE Optics and Photonics, San Diego, USA, August 2014.

Conferences

Invited Talks:

Talks:

Posters:

  • Journees Jeunes Chercheurs CNES, Toulouse, France, October 2014. 3D dark matter mapping from weak gravitational lensing. Best poster award.
  • Statistical Challenges in 21st Century Cosmology, IAU Symposium 306, Lisbon, Portugal, 25-29 May 2014. Density reconstruction from 3D lensing: Application to galaxy clusters.
  • Cosmology 2014, Rencontres de Moriond, La Thuile, Italy, 22-29 March 2014. High resolution density mapping with 3D lensing. Application to galaxy clusters.
  • Wavelet and Sparsity XV, SPIE Optics and Photonics, San Diego, USA, 26-29 August 2013. Imaging Dark Matter Through Sparsity.
  • Cosmo Probes 2013, Lausanne, Switzerland, 24-26 June 2013. Sparsity based 3d weak lensing map making.

Software

  • MRS3D: Implementation of the 3D Spherical Wavelet Transform on the Sphere in C++ with an IDL interface.
  • PRISM : Sparse recovery of the Primordial Power Spectrum implemented in C++ with an IDL interface, part of the iSAP package.
  • CosmicPy : Python package for interactive cosmology, with an embedded fast C++ library for efficient computation of Spherical Fourier-Bessel power spectra.
  • GLIMPSE : Implementation in C++ of the sparse reconstruction of 3D dark matter density from weak gravitational lensing.