Human-in-the-loop online multi-agent approach to increase trustworthiness in ML models through trust scores and data augmentation | IEEE Conference Publication | IEEE Xplore