The AIcollaboratory (aicollaboratory.org) is a data-driven framework to collect and explore data about AI results, progress and ultimately capabilities, which is being developed in the context of AI Watch, the European Commission knowledge service to monitor the development, uptake and impact of AI in Europe. The AIcollaboratory aims at integrating open data and knowledge in three different domains: an inventory of intelligent systems, a behavioural test catalogue and the measurement experiments obtained when systems take tests.
A number of examples and applications are described in the following paper: http://dmip.webs.upv.es/EPAI2020/papers/EPAI_2020_paper_3.pdf
Metamodels for Adversarial Machine Learning. Inductive programming. Model adaptation. Hierarchical and Multiclass classification. Machine Intelligence Evaluation