no code implementations • 4 Sep 2023 • Sijia Fang, Karl Rohe
These alternative models motivate a novel approach to hierarchical clustering that combines spectral techniques with the well-known Neighbor-Joining algorithm from phylogenetic reconstruction.
1 code implementation • 6 Aug 2021 • Fan Chen, Sebastien Roch, Karl Rohe, Shuqi Yu
In this situation, one could argue that the correct choice of $k$ is the number of detectable dimensions.
2 code implementations • 1 Jul 2020 • Fan Chen, Karl Rohe
Previous versions of sparse principal component analysis (PCA) have presumed that the eigen-basis (a $p \times k$ matrix) is approximately sparse.
1 code implementation • 4 Oct 2019 • Fan Chen, Yini Zhang, Karl Rohe
Using the degree-corrected stochastic block model, we study whether the PPR vector can select nodes that belong to the same block as the seed node.
1 code implementation • NeurIPS 2018 • Yilin Zhang, Karl Rohe
The second part of the paper starts from a previously proposed form of regularized spectral clustering and shows that it is related to the graph conductance on a `regularized graph'.
2 code implementations • 23 Aug 2017 • Yilin Zhang, Marie Poux-Berthe, Chris Wells, Karolina Koc-Michalska, Karl Rohe
In the Facebook data, spectral clustering without the contextualizing text information finds a mixture of (i) candidate and (ii) issue clusters.
1 code implementation • 16 Aug 2017 • Sebastien Roch, Karl Rohe
In particular, a theoretical analysis indicates that the variance of the GLS estimator is $O(n^{-1})$.
Statistics Theory Social and Information Networks Probability Statistics Theory
1 code implementation • 1 Jun 2016 • Mohammad Khabbazian, Bret Hanlon, Zoe Russek, Karl Rohe
As such, even under the ideal sampling assumptions, the performance of RDS is restricted by the underlying social network: if the network is divided into communities that are weakly connected to each other, then RDS is likely to oversample one of these communities.
Methodology
1 code implementation • 21 Aug 2015 • Juhee Cho, Donggyu Kim, Karl Rohe
In practice, the singular vectors and singular values of the low rank matrix play a pivotal role for statistical analyses and inferences.
Methodology
1 code implementation • 20 May 2015 • Karl Rohe
Under certain assumptions on the referral tree, the design effect of network sampling has a critical threshold that is a function of the referral rate $m$ and the clustering structure in the social network, represented by the second eigenvalue of the Markov transition matrix, $\lambda_2$.
no code implementations • 26 Nov 2014 • Karl Rohe
When the design matrix has orthonormal columns, "soft thresholding" the ordinary least squares (OLS) solution produces the Lasso solution [Tibshirani, 1996].
no code implementations • 8 Nov 2014 • Norbert Binkiewicz, Joshua T. Vogelstein, Karl Rohe
We utilize these node covariates to help uncover latent communities in a graph, using a modification of spectral clustering.
no code implementations • NeurIPS 2013 • Vincent Q. Vu, Juhee Cho, Jing Lei, Karl Rohe
We propose a novel convex relaxation of sparse principal subspace estimation based on the convex hull of rank-$d$ projection matrices (the Fantope).
no code implementations • NeurIPS 2013 • Tai Qin, Karl Rohe
Spectral clustering is a fast and popular algorithm for finding clusters in networks.
no code implementations • 8 Jul 2013 • Karl Rohe, Tai Qin
The only constraint on the ambient graph is that it is large and sparse--it could be generated at random or by an adversary--suggesting a theoretical explanation for the robust empirical performance of local clustering algorithms.
no code implementations • 10 Apr 2012 • Karl Rohe, Tai Qin, Bin Yu
In each example, a small subset of nodes have persistent asymmetries; these nodes send edges with one cluster, but receive edges with another cluster.