際際滷

際際滷Share a Scribd company logo

The ways of node embedding

?
?
S
SEMINARGROOT

Node embedding techniques learn vector representations of nodes in a graph that can be used for downstream machine learning tasks like classification, clustering, and link prediction. DeepWalk uses random walks to generate sequences of nodes that are treated similarly to sentences, and learns embeddings by predicting nodes using their neighbors, like word2vec. It does not incorporate node features or labels. Node2vec extends DeepWalk by introducing a biased random walk to learn embeddings, addressing some limitations of DeepWalk while maintaining scalability.

1 of 17
Download to read offline
The ways of node embedding
Node Embedding ? Goal : Efficient feature learning for ML ? Learned vectors can be used for ? Classification : SVM, Logistic Regression ? Clustering : K-means ? Link Prediction
Node Embedding ??? ?, ? 「 ???(?!, ?")
Node Embedding ? How to define Node Similarity? Common Neighbors : |? ? ” ? ? | Adamic-Adar Index : ‘ !（ # $ ”# & ' ()* |# ! | Jaccard Similarity : # $ ”# & # $ “# &
Node Embedding ? Why not?? ? Unseen node -> how to embed? ? ?(|?|) parameters needed ? Cannot incorporate node features ? GCN
Neighborhood Aggregation
Neighborhood Aggregation ?! " = ?! ?! # = ?(?# ' $（& ! ?$ #'( |? ? | + ?#?! #'( )
Node embedding We now introduce the ways of node embedding using idea of word embedding
DeepWalk : Online Learning of Social Representations ?? ??? ????? ____ ? ??? ?? ??? ??? ??? ???? ?? ??? ????? ????.(CBOW) ? = ??????$ ??? ??, ? , ??? ???, ? , ???(???, ?)
DeepWalk : Online Learning of Social Representations ___ ??? ___ ??? ??? ?? ??? ???? ????? ????? ????.(Skip-Gram) ? = ??????$ ??? ???, ?
DeepWalk : Online Learning of Social Representations How can we make a sentence of nodes? Random Walk Neighborhood preserving likelihood Encoder-decoder structure
DeepWalk : Online Learning of Social Representations
DeepWalk : Online Learning of Social Representations ???, SkipGram?? ????? ??? ?? V??? ?? ???? Hierarchical Softmax h? W? ??? softmax : O(N) -> infeasible O(logN) ??? ?????
DeepWalk : Online Learning of Social Representations Negative Sampling ??? ?? ?? : |V| -> K
DeepWalk : Online Learning of Social Representations Pros : 1. Unseen data? ??? ???? 2. Node/Edge feature ?? ??. 3. Node label ?? ?? Cons : 1. Unseen edge? ???? 2. Local information ?? 3. ???? ??
Node2vec: Scalable Feature Learning for Networks
The ways of node embedding

More Related Content

The ways of node embedding

  • 2. Node Embedding ? Goal : Efficient feature learning for ML ? Learned vectors can be used for ? Classification : SVM, Logistic Regression ? Clustering : K-means ? Link Prediction
  • 3. Node Embedding ??? ?, ? 「 ???(?!, ?")
  • 4. Node Embedding ? How to define Node Similarity? Common Neighbors : |? ? ” ? ? | Adamic-Adar Index : ‘ !（ # $ ”# & ' ()* |# ! | Jaccard Similarity : # $ ”# & # $ “# &
  • 5. Node Embedding ? Why not?? ? Unseen node -> how to embed? ? ?(|?|) parameters needed ? Cannot incorporate node features ? GCN
  • 7. Neighborhood Aggregation ?! " = ?! ?! # = ?(?# ' $（& ! ?$ #'( |? ? | + ?#?! #'( )
  • 8. Node embedding We now introduce the ways of node embedding using idea of word embedding
  • 9. DeepWalk : Online Learning of Social Representations ?? ??? ????? ____ ? ??? ?? ??? ??? ??? ???? ?? ??? ????? ????.(CBOW) ? = ??????$ ??? ??, ? , ??? ???, ? , ???(???, ?)
  • 10. DeepWalk : Online Learning of Social Representations ___ ??? ___ ??? ??? ?? ??? ???? ????? ????? ????.(Skip-Gram) ? = ??????$ ??? ???, ?
  • 11. DeepWalk : Online Learning of Social Representations How can we make a sentence of nodes? Random Walk Neighborhood preserving likelihood Encoder-decoder structure
  • 12. DeepWalk : Online Learning of Social Representations
  • 13. DeepWalk : Online Learning of Social Representations ???, SkipGram?? ????? ??? ?? V??? ?? ???? Hierarchical Softmax h? W? ??? softmax : O(N) -> infeasible O(logN) ??? ?????
  • 14. DeepWalk : Online Learning of Social Representations Negative Sampling ??? ?? ?? : |V| -> K
  • 15. DeepWalk : Online Learning of Social Representations Pros : 1. Unseen data? ??? ???? 2. Node/Edge feature ?? ??. 3. Node label ?? ?? Cons : 1. Unseen edge? ???? 2. Local information ?? 3. ???? ??
  • 16. Node2vec: Scalable Feature Learning for Networks