I am a fourth year student at Ecole Normale Superieure de Paris, currently visiting Stanford University under the supervision of Prof. Andrea Montanari. Before this year, I have earned a M.Sc. in Theoretical Physics (ICFP, ENS) and both a B.Sc. in Mathematics and a B.Sc. in Physics at ENS de Paris.
My research interest lies broadly at the intersection of statistical physics, information theory, computation and probability theory. My goal is to understand what is makes a problem hard or easy. I am particularly interested in finding optimal algorithms, mapping the tradeoffs between sample complexity, error probability and computation time, and studying computational phase transitions (where a small change of parameters causes a sharp change in the computational complexity of the problem).
Currently, I am working on high-dimensional statistical estimation, SDP problems and non-convex optimization. The applications range from Graphical Models selection, to Max-Cut and phase synchronization.
Concentration to Zero Bit-Error Probability for Regular LDPC Codes on the Binary Symmetric Channel: Proof by Loop Calculus. Marc Vuffray and Theodor Misiakiewicz. 53rd Annual Allerton Conference on Communication, Control, and Computing (Allerton, 2015).
Efficient reconstruction of transmission probabilities in a spreading process from partial observations. Andrey Y Lokhov and Theodor Misiakiewicz. Submitted.
See the Research section of this website.
Department of Electrical Engineering
Stanford, CA 94305
theodor.misiakiewicz (at) ens.fr