Myriam Schönenberger

I'm finishing my Master's at the Kim Jaechul Graduate School of AI at KAIST this summer (August 2026), where I worked in the AI & Probabilistic Reasoning (AIPR) Lab with Kee-Eung Kim. My thesis develops a framework for discovering control symmetries directly from transition data, by identifying group actions whose orbits remain tangent to the manifold of valid state–action–next-state transitions.

More broadly, I'm interested in how structure — symmetries, invariances, and the geometry of dynamics — can make reinforcement learning agents more sample-efficient, robust, and safe to deploy in the real world.

Before AI, I was a particle physicist at the Institute for Particle Physics and Astrophysics (IPA) at ETH Zürich, advised by Rainer Wallny. I searched for supersymmetry in the fully hadronic and Higgs-to-diphoton final states with the CMS experiment at the LHC, summarized in my PhD thesis.

CV / Scholar / Github

Publications

Listed below are publications where I contributed significantly.

	Search for supersymmetry using Higgs boson to diphoton decays at √s = 13 TeV CMS Collaboration CMS, JHEP, 11, 2019, 109 10.1007/JHEP11(2019)109 1908.08500
	Search for new phenomena with the MT2 variable in the all-hadronic final state produced in proton–proton collisions at √s = 13 TeV CMS Collaboration CMS, Eur. Phys. J., C77, 2017, 10, 710 10.1140/epjc/s10052-017-5267-x 1705.04650
	Search for new physics with the MT2 variable in all-jets final states produced in pp collisions at √s = 13 TeV CMS Collaboration CMS, JHEP, 10, 2016, 006 10.1007/JHEP10(2016)006 1603.04053
	Test beam results with a sampling calorimeter of cerium fluoride scintillating crystals and tungsten absorber plates for calorimetry at the HL-LHC R. Becker and others Nucl. Instrum. Meth., A824, 2016, 681-683 10.1016/j.nima.2015.09.052
	Searches for strong production of supersymmetry in the fully hadronic final states at CMS Schoenenberger, Myriam PoS, EPS-HEP2017, 2017, 336 10.22323/1.314.0336

Original source code by Jon Barron.