<ul><li>Research and experiments have shown that Group Equivariant Non-Expansive Operators (GENEOs) in Machine Learning and Artificial Intelligence can be used to develop explainable models.</li><li>A study focused on GENEOnet, a GENEO network used in computational biochemistry, has conducted statistical analysis and experiments to confirm its explainability and trustworthiness.</li><li>The sensitivity analysis of GENEOnet's parameters showed their significance, and GENEOnet displayed a higher proportion of equivariance compared to other methods.</li><li>GENEOnet also demonstrated robustness to perturbations arising from molecular dynamics. These findings confirm the trustworthiness and beneficial use of GENEOs in the context of Trustworthy Artificial Intelligence.</li></ul>

GENEOnet: Statistical analysis supporting explainability and trustworthiness

Discover more