표·그림DB
Fig. 3.30. Manifold of the embedding vector spaces at each stage of the proposed representation learning procedures. (a) shows the given embedding vectors, (b) shows the pretrained embedding vectors (z̄pre) obtained by self-supervised learning, and (c) shows the embedding vectors trained by multi-label soft-silhouette deep clustering. The samples in the same cluster are represented with the same color. The clustering results are evaluated by silhouette-coefficient score(s).
표·그림명
Fig. 3.30. Manifold of the embedding vector spaces at each stage of the proposed representation learning procedures. (a) shows the given embedding vectors, (b) shows the pretrained embedding vectors (z̄pre) obtained by self-supervised learning, and (c) shows the embedding vectors trained by multi-label soft-silhouette deep clustering. The samples in the same cluster are represented with the same color. The clustering results are evaluated by silhouette-coefficient score(s).