Pierre-Cyril Aubin-Frankowski

Papers

Total Citations

H-Index

About

Pierre-Cyril Aubin-Frankowski is a researcher at the forefront of machine learning and optimization, with a core focus on shape-constrained learning and kernel methods. His most-cited work, "Handling Hard Affine SDP Shape Constraints in RKHSs" (2021), tackles the challenging problem of enforcing hard shape constraints—such as non-negativity, monotonicity, and convexity—over continuous domains in Reproducing Kernel Hilbert Spaces. This contribution is pivotal for applications requiring rigorous adherence to physical or economic principles, bridging the gap between theoretical guarantees and practical predictive modeling. With over 5 citations, his work has already influenced researchers in statistics and optimization. Aubin-Frankowski’s research also extends to control theory and dynamical systems, where he develops algorithms that integrate constraints into learning pipelines. His achievements include advancing the understanding of semidefinite programming for shape constraints, offering a robust framework for safe and interpretable AI. For students and researchers, his work exemplifies how mathematical rigor can drive reliable machine learning in high-stakes domains.