High performing machine learning for novel catalyst design (MLNovCat)
Main funder
Funder's project number: 351579
Funds granted by main funder (€)
- 354 112,00
Funding program
Project timetable
Project start date: 01/01/2022
Project end date: 31/12/2024
Summary
Cleanly produced hydrogen, which can be produced through water electrocatalysis, is crucial for achieving a low-carbon society. Novel, next-generation catalysts for this reaction can be based on small monolayer-protected clusters (MPCs), which contain multiple tunable properties. To speed up their design, high performing and reliable data-driven methods utilizing graphics processing units (GPU) should be applied. In the project, a new concept for the design of catalysts is created, which can replace the conventional trial-and-error experimental laboratory work. The consortium for the project is interdisciplinary, consisting of three groups at the University of Jyväskylä that have demonstrated complementary expertise in the computational catalysis, materials science, and computational science.
Principal Investigator
Other persons related to this project (JYU)
Primary responsible unit
Other responsible units
Follow-up groups
Related publications and other outputs
- Additive autoencoder for dimension estimation (2023) Kärkkäinen, Tommi; et al.; A1; OA
- Feature selection for distance-based regression : An umbrella review and a one-shot wrapper (2023) Linja, Joakim; et al.; A1; OA
- Graphs and Kernelized Learning Applied to Interactions of Hydrogen with Doped Gold Nanoparticle Electrocatalysts (2023) Pihlajamäki, Antti; et al.; A1; OA
- More ConvNets in the 2020s : Scaling up Kernels Beyond 51x51 using Sparsity (2023) Liu, Shiwei; et al.; D3; OA