Random projection tree similarity metric for SpectralNet
Not on the map. Why?
Abstract (arXiv)
SpectralNet is a graph clustering method that uses neural network to find an embedding that separates the data. So far it was only used with k-nn graphs, which are usually constructed using a distance metric (e.g., Euclidean distance). k-nn graphs restrict the points to have a fixed number of neighbors regardless of the local statistics around them. We proposed a new SpectralNet similarity metric based on random projection trees (rpTrees). Our experiments revealed that SpectralNet produces better clustering accuracy using rpTree similarity metric compared to k-nn graph with a distance metric. Also, we found out that rpTree parameters do not affect the clustering accuracy. These parameters include the leaf size and the selection of projection direction. It is computationally efficient to keep the leaf size in order of log(n), and project the points onto a random direction instead of trying to find the direction with the maximum dispersion.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 9 of its 32 references that are in the map. First posted on arXiv 25 Feb 2023.
Builds on
9 of 32 references.
References · 9
- 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
- 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,898
- 2019Learning Discrete Structures for Graph Neural Networkscited by 231
- 2018SpectralNet: Spectral Clustering using Deep Neural Networkscited by 94
- 2018K-nearest Neighbor Search by Random Projection Forestscited by 10
- 2019Spectral Clustering via Ensemble Deep Autoencoder Learning (SC-EDAE)cited by 10
- 2023Refining a k-nearest neighbor graph for a computationally efficient spectral clusteringcited by 6
- 2021Variational Co-embedding Learning for Attributed Network Clusteringcited by 4
- 2019Similarity Kernel and Clustering via Random Projection Forestscited by 2
Built on by
Not cited yet.