This paper proposes a unified framework for attributed network embedding, Attri2vec, that learns node-embeddings by discovering a latent node attribute subspace via a network structure guided transformation performed on the original attribute space. The resultant latent subspace can respect network structure in a more consistent way towards learning high-quality node representations. Unlike other structure-aware node-embedding methods, this approach does not rely on message passing mechanism and hence very scalable.
🚀 The feature, motivation and pitch
I would like to implement the paper Attributed Network Embedding via Subspace Discovery by Zhang et al.
This paper proposes a unified framework for attributed network embedding, Attri2vec, that learns node-embeddings by discovering a latent node attribute subspace via a network structure guided transformation performed on the original attribute space. The resultant latent subspace can respect network structure in a more consistent way towards learning high-quality node representations. Unlike other structure-aware node-embedding methods, this approach does not rely on message passing mechanism and hence very scalable.
Alternatives
Attri2vec is currently implemented in Stellargraph, which is based on Tensorflow.
Additional context
No response