Benchmarking Methods for Science
A benchmark is a standardised validation framework that allows for the direct comparison of different models, and consequently theories, that address the same research problem. Traditionally, participants are invited to submit their models to a benchmark with a specific focus, which is commonly referred to as "organising a challenge" and their submissions are assessed using a predefined set of evaluation criteria. Benchmark challenges have the potential to be highly collaborative, as they encourage numerous researchers from different fields with diverse backgrounds and expertise to work together on the same research problem.
(PDF) Accelerating progress in the social sciences: the potential of benchmarks
During the SICSS-ODISSEI Summer School in June 2023, a Fertility Prediction Challenge was organised. This Challenge functioned as a pilot to test the first version of the software service and to gather feedback from stakeholders on how to improve the software. Visit the Fertility Prediction Challenge GitHub page for more information about the challenge.
Presentation about the summer school pilot at the ODISSEI conference 2023
Presentation about the summer school pilot at the ODISSEI conference 2023
| Eyra Leap B.V. | Contact | LinkedIn | 🌐 eyra.co | KVK 77867513 | Saturnusstraat 14 - Unit 4.13 - 2516 AH The Hague |