국립한국교통대학교 AMI Lab.
A learning paradigm for extreme low-label settings where labeled data is fewer than the number of classes.
MORE VIEWAnalysis of reliability and collapse of learning systems under uncertainty across data, architecture, and training.
MORE VIEWA learning framework for reliable time-series anomaly detection in resource-constrained edge environments.
MORE VIEWAn analytical framework for automatically collecting and refining product data to intelligently recommend competing products.
MORE VIEW