A Notation Table, M or M

We summarize several representative works under this unified framework by listing the exact forms of two key matrices S and Λ in Table 5. Gaussian Random Field (GRF) [85] is the first to propose the idea of label propagation by using the combinatorial Laplacian matrix as the smoothing matrix. It also tolerates no prediction errors on the labeled nodes, so the trade-off parameter can be set to infinity. Local and Global Consistency (LGC) [82] relaxes the GRF method by eliminating the restriction of zero empirical risk on labeled nodes and exploits the normalized Laplacian matrix for smoothing instead.