Randomized classifiers have been shown to provide a promising approach for achieving certified robustness against adversarial attacks in deep learning.

However, the landscape of6 synthetic data research has been fragmented due to the large number of data7 modalities(e.g.,tabulardata,timeseriesdata,images,etc.) andvarioususecases8 (e.g., privacy, fairness, data augmentation, etc.). Beyond benchmarking, it also offers a single access point to a diverse range of15 cutting-edge data generators.