Marcus Carlsson: Non-convex optimisation

Välkommen till en föreläsning i seminarieserien i matematik.


Marcus Carlsson


Non-convex optimisation


A common technique for solving ill-posed inverse problems is to include some sparsity constraint, and pose it as a convex optimization problem, as is done e.g. in compressive sensing. The corresponding functional to be minimized often includes an l2 data fidelity term plus a convex term forcing sparsity. However, for many applications a non-convex term would be more suitable, although this is discarded since it leads to issues with algorithm convergence, local minima etc. I will introduce a new transform to convexify non-convex functionals of the above type. Applications include low rank approximation, multi-dimensional frequency estimation, phase retrieval, GPS-positioning, among others. I will discuss some of these if time allows.


Seminariet är en del av den seminarieserie som forskningsämnet matematik arrangerar regelbundet, med såväl inbjudna som egna forskare. Här kan du se hela seminarieserien.

Rum D1172, hus D, Växjö Joachim Toft Lägg till i din kalender