Discover and Align Taxonomic Context Priors for Open-world Semi-Supervised Learning (Supplementary Materials)

In this section, we provide another interpretation of TIDA, which can provide more insights into the nature of the learned structure priors. TIDA intrinsically: (a) builds hierarchical vMF distributions to cluster samples and discovery taxonomic context by optimizing Eq. VI; (b) applies the consistency constraint on the hierarchical vMF distributions to build communication and alignment across taxonomic context as Eq.