Benchmark dataset for diversity-aware batch-mode active learning using Hill's anisotropic yield criterion

This collection contains FAIR-described data objects generated for benchmarkingdiversity-aware batch-mode active learning in six-dimensional stress space, usingHill’s anisotropic yield criterion as the analytical oracle for yield-onsetdetermination. Each data object represents one independent active-learning run andcontains the corresponding yield-onset stress points. The collection spans differentbatch sizes and repeated runs. Within each data object, the relevant metadata andcontextual information are stored to support the FAIR principles of findability,accessibility, interoperability, and reusability.