TY - JOUR A1 - Plock, Matthias A1 - Burger, Sven A1 - Schneider, Philipp-Immanuel T1 - Recent advances in Bayesian optimization with applications to parameter reconstruction in optical nano-metrology JF - Proc. SPIE Y1 - 2021 U6 - https://doi.org/10.1117/12.2592266 VL - 11783 SP - 117830J ER - TY - JOUR A1 - Plock, Matthias A1 - Andrle, Kas A1 - Burger, Sven A1 - Schneider, Philipp-Immanuel T1 - Bayesian Target-Vector Optimization for Efficient Parameter Reconstruction JF - Adv. Theory Simul. Y1 - 2022 U6 - https://doi.org/10.1002/adts.202200112 VL - 5 SP - 2200112 ER - TY - JOUR A1 - Rickert, Lucas A1 - Betz, Fridtjof A1 - Plock, Matthias A1 - Burger, Sven A1 - Heindel, Tobias T1 - High-performance designs for fiber-pigtailed quantum-light sources based on quantum dots in electrically-controlled circular Bragg gratings JF - Opt. Express Y1 - 2023 U6 - https://doi.org/10.1364/OE.486060 VL - 31 SP - 14750 ER - TY - JOUR A1 - Bopp, Julian M. A1 - Plock, Matthias A1 - Turan, Tim A1 - Pieplow, Gregor A1 - Burger, Sven A1 - Schröder, Tim T1 - Sawfish Photonic Crystal Cavity for Near-Unity Emitter-to-Fiber Interfacing in Quantum Network Applications JF - Adv. Opt. Mater. Y1 - 2024 U6 - https://doi.org/10.1002/adom.202301286 VL - 12 SP - 2301286 ER - TY - JOUR A1 - Plock, Matthias A1 - Binkowski, Felix A1 - Zschiedrich, Lin A1 - Schneider, Phillip-Immanuel A1 - Burger, Sven T1 - Fabrication uncertainty guided design optimization of a photonic crystal cavity by using Gaussian processes JF - J. Opt. Soc. Am. B Y1 - 2024 U6 - https://doi.org/10.1364/JOSAB.505767 VL - 41 SP - 850 ER - TY - JOUR A1 - Sekulic, Ivan A1 - Schaible, Jonas A1 - Müller, Gabriel A1 - Plock, Matthias A1 - Burger, Sven A1 - Martínez-Lahuerta, Víctor José A1 - Gaaloul, Naceur A1 - Schneider, Philipp-Immanuel T1 - Physics-informed Bayesian optimization of expensive-to-evaluate black-box functions JF - Mach. Learn. Sci. Technol. N2 - Abstract Bayesian optimization with Gaussian process surrogates is a popular approach for optimizing expensive-to-evaluate functions in terms of time, energy, or computational resources. Typically, a Gaussian process models a scalar objective derived from observed data. However, in many real-world applications, the objective is a combination of multiple outputs from physical experiments or simulations. Converting these multidimensional observations into a single scalar can lead to information loss, slowing convergence and yielding suboptimal results. To address this, we propose to use multi-output Gaussian processes to learn the full vector of observations directly, before mapping them to the scalar objective via an inexpensive analytical function. This physics-informed approach retains more information from the underlying physical processes, improving surrogate model accuracy. As a result, the approach accelerates optimization and produces better final designs compared to standard implementations. Y1 - 2025 U6 - https://doi.org/10.1088/2632-2153/ae1f5f VL - 6 SP - 040503 PB - IOP Publishing ER - TY - JOUR A1 - Binkowski, Felix A1 - Koulas-Simos, Aris A1 - Betz, Fridtjof A1 - Plock, Matthias A1 - Sekulic, Ivan A1 - Manley, Phillip A1 - Hammerschmidt, Martin A1 - Schneider, Philipp-Immanuel A1 - Zschiedrich, Lin A1 - Munkhbat, Battulga A1 - Reitzenstein, Stephan A1 - Burger, Sven T1 - High Purcell enhancement in all-TMDC nanobeam resonator designs with active monolayers for nanolasers JF - Phys. Rev. B Y1 - 2025 U6 - https://doi.org/10.1103/nxh9-dhvx VL - 112 SP - 235410 ER -