|
bansal-vansh.github.io
|
Vansh Bansal
|
|
syamantakk.github.io
|
Syamantak Kumar
|
|
sites.google.com
|
Tuan Pham
|
|
akhiljalan.github.io
|
Akhil Jalan
|
|
sites.google.com
|
Saptarshi Roy
|
|
rslunde.github.io
|
Robert Lunde
|
|
qiaohuilin.github.io
|
Qiaohui Lin
|
|
sites.google.com
|
Prateek Srivastava
|
|
cs.utexas.edu
|
Xueyu Mao
|
|
boweiyan.github.io
|
Bowei Yan
|
|
arxiv.org
|
<span style="font-weight: bold;">Arxiv</span>
|
|
arxiv.org
|
<span style="font-weight: bold;">
Combinatorial Sparse PCA Beyond the Spiked Identity Model </span>
|
|
arxiv.org
|
<span style="font-weight: bold;">
Dimension-free Score Matching and Time Bootstrapping for Diffusion Models </span>
|
|
arxiv.org
|
<span style="font-weight: bold;">
Low-precision streaming PCA </span>
|
|
arxiv.org
|
<span style="font-weight: bold;">
Spike-and-Slab Posterior Sampling in High Dimensions </span>
|
|
pubsonline.informs.org
|
<span style="font-weight: bold;">
Optimal Transfer Learning for Missing Not-at-Random Matrix Completion </span>
|
|
pubsonline.informs.org
|
<span style="font-weight: bold;">
Incentive-Aware Models of Financial Networks </span>
|
|
arxiv.org
|
<span style="font-weight: bold;">
Oja's algorithm for Sparse PCA </span>
|
|
arxiv.org
|
<span style="font-weight: bold;">
On Differentially Private U-statistics </span>
|
|
arxiv.org
|
<span style="font-weight: bold;">
Transfer Learning for Latent Variable Network Models </span>
|
|
arxiv.org
|
<span style="font-weight: bold;">
Keep or toss? A nonparametric score to evaluate solutions for noisy ICA </span>
|
|
arxiv.org
|
<span style="font-weight: bold;">
Streaming PCA for Markovian Data </span>
|
|
www3.stat.sinica.edu.tw
|
<span style="font-weight: bold;">
A Unified Framework for Tuning Hyperparameters in Clustering Problems
</span>
|
|
openreview.net
|
<span style="font-weight: bold;">Bootstrapping the error of Oja’s algorithm.
</span>
|
|
pnas.org
|
Two provably consistent divide and conquer clustering algorithms for large networks
|
|
arxiv.org
|
<span style="font-weight: bold;">When random initializations help: a study of variational inference for community detection
</span>
|
|
arxiv.org
|
<span style="font-weight: bold;">On the Theoretical Properties of the Network Jackknife
</span>
|
|
proceedings.mlr.press
|
<span style="font-weight: bold;">On hyperparameter tuning in general clustering problems
</span>
|
|
arxiv.org
|
<span style="font-weight: bold;">A Theoretical Case Study of Structured Variational Inference for Community Detection
</span>
|
|
arxiv.org
|
<span style="font-weight: bold;">Hierarchical Community Detection by Recursive Partitioning
</span>
|
|
arxiv.org
|
<span style="font-weight: bold;">Estimating Mixed Memberships with Sharp Eigenvector Deviations
</span>
|
|
arxiv.org
|
<span style="font-weight: bold;">Provable Estimation of the Number of Blocks in Block Models
</span>
|
|
arxiv.org
|
<span style="font-weight: bold;">On clustering network-valued data.
</span>
|
|
arxiv.org
|
<span style="font-weight: bold;">Statistical Convergence Analysis of Gradient EM on multi-component Gaussian Mixture Models.
</span>
|
|
cs.utexas.edu
|
<span style="font-weight: bold;">On Mixed Memberships and Symmetric Nonnegative Matrix Factorizations.
</span>
|
|
cs.utexas.edu
|
<span style="font-weight: bold;">[Supplementary]</span>
|
|
cs.utexas.edu
|
<span style="font-weight: bold;">[code]</span>
|
|
papers.nips.cc
|
<span style="font-weight: bold;">On Robustness of Kernel Clustering.
</span>
|
|
arxiv.org
|
<span style="font-weight: bold;">[arxiv]</span>
|
|
boweiyan.github.io
|
<span style="font-weight: bold;">[code]</span>
|
|
cacm.acm.org
|
<span style="font-weight: bold;">
Answering Enumeration Queries with the Crowd.
</span>
|
|
faculty.mccombs.utexas.edu
|
<span style="font-weight: bold;">The Consistency of Common Neighbors for Link Prediction in Stochastic Blockmodels.
</span>
|
|
faculty.mccombs.utexas.edu
|
<span style="font-weight: bold;">[Supplementary]</span>
|
|
web.eecs.umich.edu
|
<span style="font-weight: bold;">Scaling Up Crowd-Sourcing to Very Large Datasets: A Case for Active Learning.
</span>
|
|
arxiv.org
|
<span style="font-weight: bold;">Role of normalization in spectral clustering.</span>
|
|
arxiv.org
|
<span style="font-weight: bold;">Hypothesis Testing for Automated Community Detection in Networks.</span>
|
|
amplab.cs.berkeley.edu
|
<span style="font-weight: bold;">Crowdsourced Enumeration Queries.</span>
|
|
cs.berkeley.edu
|
<span style="font-weight: bold;">Nonparametric Link Prediction in Dynamic Networks.</span>
|
|
cs.berkeley.edu
|
[Appendix]
|
|
arxiv.org
|
<span style="font-weight: bold;">Nonparametric Link Prediction in Large Scale Dynamic Networks.</span>
|
|
cs.berkeley.edu
|
<span style="font-weight: bold;">Big Data Bootstrap.</span>
|
|
rss.onlinelibrary.wiley.com
|
<span style="font-weight: bold;">A Scalable Bootstrap for Massive Data.</span>
|
|
faculty.mccombs.utexas.edu
|
<span style="font-weight: bold;">Theoretical Justification of Popular Link Prediction Heuristics.</span>
|
|
cs.cmu.edu
|
<span style="font-weight: bold;">Here</span>
|
|
dl.acm.org
|
<span style="font-weight: bold;">Fast Nearest-neighbor Search in Disk-resident Graphs.</span>
|
|
cs.cmu.edu
|
<span style="font-weight: bold;">here.</span> <br>
<br>
<br>
<br>
<span style="color: rgb(0, 0, 0);"> </span>
|
|
dl.acm.org
|
<span style="font-weight: bold;">Fast Dynamic Reranking in Large Graphs. </span>
|
|
cs.cmu.edu
|
<span style="font-weight: bold;">Here</span>
|
|
autonlab.org
|
<span style="color: rgb(0, 0, 0);"> </span><span style="font-weight: bold;"><a style="color: rgb(0, 102, 179); font-weight: bold;" href="http://icml2008.cs.helsinki.fi/papers/565.pdf">Fast Incremental Proximity Search in Large Graphs.
</a> (International Conference on Machine Learning, ICML 2008)
<!---[Slightly revised from the ICML camera-ready version]<br />
--->
<br>(P. Sarkar, A. W. Moore, and A. Prakash)
<br><br>
- In this paper we combine random sampling with
deterministic neighborhood expansion schemes to compute approximate
nearest neighbors of a query under hitting and commute times.
Resulting algorithms can process queries on the fly without any
preprocessing. This is of crucial importance for graphs which change
quickly over time. <a href="icml_talk.ppt">Here</a> is the presentation.
<br>
<br>
<br>
<span style="color: rgb(0, 0, 0);"> </span><a style="color: rgb(0, 102, 179); font-weight: bold;" href="https://arxiv.org/pdf/1206.5259.pdf">A Tractable Approach to Finding Closest Truncated-commute-time Neighbors in Large Graphs.
</a>(Uncertainty in Artificial Intelligence, UAI 2007)<br style="font-weight: bold;">
(P. Sarkar and A. W. Moore)
<br><br>
- We present GRANCH, an algorithm to compute all
interesting pairs of approximate nearest neighbors in truncated
commute times in a graph, without
computing it between all pairs. Our algorithm provably prunes away
uninteresting pairs of nodes in the graph, and as a result
quickly finds the "potential nearest neighbors".<br>
<br>
<br>
<span style="color: rgb(0, 0, 0);"> </span><a style="color: rgb(0, 102, 179); font-weight: bold;" href="http://proceedings.mlr.press/v2/sarkar07a/sarkar07a.pdf">A Latent Space Approach to Dynamic Embedding of Co-occurrence Data.
</a>(International Conference on Artificial Intelligence and Statistics, AISTATS 2007)<br>
(P. Sarkar, S. Siddiqi, and G. Gordon)
<br><br>
- We propose a graphical model to compute dynamic
embeddings of co-occurrence data. We show how to make inference
tractable by using Kalman Filters, which also provides distributional
information of the embedding inferred. <br>
<span style="color: rgb(0, 0, 0);"><br>
<br>
</span><a style="color: rgb(0, 102, 179); font-weight: bold;" href="NIPS2005_0724.pdf">Dynamic Social Network Analysis using Latent Space Models.
</a>(Advances in Neural Information Processing Systems, NIPS 2005)<br>
(P. Sarkar and A. W. Moore)
<br><br>
- We present a dynamic model for social networks changing
over time. We combine a new dynamic multidimentional scaling algorithm
with a tractable local optimization technique for inference. Both of
these algorithms are highly efficient (sub-quadratic in size of the
social network).<br>
<br>
- </span>
|
|
icml2008.cs.helsinki.fi
|
Fast Incremental Proximity Search in Large Graphs.
|
|
arxiv.org
|
A Tractable Approach to Finding Closest Truncated-commute-time Neighbors in Large Graphs.
|
|
proceedings.mlr.press
|
A Latent Space Approach to Dynamic Embedding of Co-occurrence Data.
|
|
ml.cmu.edu
|
[<i>Extended version</i>:]
|
WordPress