Sadegh Jokar, Data Scientist at GameDuell
Estimating Representational Dissimilarity Measures‎‎

The talk discusses some theoretical and numerical results for compressive sensing (CS). The first part is concerned different sparsity measures such as restricted isometry property (RIP), null space property (NSP), mutual incoherence, spark and their connections and how these measures behave in the case where the matrix is the Kronecker product of matrices. Then a framework for sparse representation of solutions of PDEs using the notion of multilevel dictionaries and Galerkin approach will be presented. The PDE is discretized in the Galerkin framework and solved using a recursive frame refinement procedure, that uses so called orthogonal matching pursuit and basis pursuit to find a good approximation to the sparse solution on a given refinement level.

Date & Time: 
‎Saturday‎, December 31‎, ‎2016, ‎‎11:00 - 12:00
‎Room 221‎, ‎Dept‎. ‎Math‎. ‎Sci.‎, ‎Sharif University of Technology‎