Sujay Sanghavi - Papers by Year

Recent

• Interaction Hard Thresholding: Consistent Sparse Quadratic Regression in Sub-quadratic Time and Space
  Shuo Yang, Yanyao Shen, Sujay Sanghavi
  NeurIPS 2019

• Blocking Bandits
  Soumya Basu, Rajat Sen, Sujay Sanghavi, Sanjay Shakkottai
  NeurIPS 2019

• Learning Distributions Generated by One-Layer ReLU Networks
  Shanshan Wu, Alexandros Dimakis, Sujay Sanghavi
  NeurIPS 2019

• Sparse Logistic Regression Learns All Discrete Pairwise Graphical Models
  Shanshan Wu, Sujay Sanghavi, Alexandros Dimakis
  NeurIPS 2019

• Iterative Least Trimmed Squares for Mixed Linear Regression
  Yanyao Shen, Sujay Sanghavi
  NeurIPS 2019

• PruneTrain: Fast Neural Network Training by Dynamic Sparse Model Reconfiguration
  Sangkug Lym, Esha Choukse, Siavash Zangeneh, Wei Wen, Sujay Sanghavi, Mattan Erez
  Published in Supercomputing (SC) 2019

• Learning a Compressed Sensing Measurement Matrix via Gradient Unrolling
  S. Wu, A. Dimakis, S. Sanghavi, F. Yu, D. Holtmann-Rice, D. Storcheus, A. Rostamizadeh, S. Kumar
  ICML 2019

• Learning with Bad Training Data via Iterative Trimmed Loss Minimization
  Y. Shen, S. Sanghavi
  ICML 2019

• Online Collaborative Filtering of Graphs
  S. Banerjee, S. Sanghavi and S. Shakkottai
  Operations Research

• Matrix Completion with Column Manipulation: Near-Optimal Sample-Robustness-Rank Tradeoffs
  Y. Chen, H. Xu, C. Caramanis and S. Sanghavi
  IEEE Transactions on Information Theory ( Volume: 62 , Issue: 1 , Jan. 2016 )

• Single Pass PCA of Matrix Products
  S. Wu, S. Bhojanapalli, S. Sanghavi, A. Dimakis
  NIPS 2016

• Normalized Spectral Map Synchronization
  Y. Shen, Q. Huang, N. Srebro, S. Sanghavi
  NIPS 2016

• Preference Completion: Large-scale Collaborative Ranking from Pairwise Comparisons
  D. Park, J. Neeman, J. Zhang, S. Sanghavi, I. Dhillon
  ICML 2015

• The Local Convexity of Solving Quadratic Equations
  C. White, R. Ward, S. Sanghavi
  Results in Mathematics (2016): 1-40

2015

• Serving content with unknown demand: the high-dimensional regime
  S. Moharir, J. Ghaderi, S. Sanghavi and S. Shakkottai
  Queueing systems vol. 81 (2), 2015.

• Convergence Rates of Active Learning for Maximum Likelihood Estimation
  K. Chaudhuri, S. Kakade, P. Netrapalli, S. Sanghavi
  Neural Information Processing Systems (NIPS) 2015

• A New Sampling Technique for Tensors
  S. Bhojanapalli, S. Sanghavi
  preprint

• Tighter Low-rank Approximation via Sampling the Leveraged Element
  S. Bhojanapalli, P. Jain, S. Sanghavi
  ACM-SIAM Symposium on Discrete Algorithms (SODA) 2015

• Completing Any Matrix, Provably
  Y. Chen, S. Bhojanapalli, S. Sanghavi, R. Ward
  to appear in JMLR

2014

• Non-convex Robust PCA
  P. Netrapalli, U. Niranjan, S. Sanghavi, A. Anandkumar, P. Jain
  Neural Information Processing Systems (NIPS) 2014

• Greedy Subspace Clustering
  D. Park, C. Caramanis, S. Sanghavi
  Neural Information Processing Systems (NIPS) 2014

• Alternating Minimization for Mixed Linear Regression
  X. Yi, C. Caramanis, S. Sanghavi
  International Conference on Machine Learning (ICML) 2014

• Coherent Matrix Completion
  Y. Chen, S. Bhojanapalli, S. Sanghavi, R. Ward
  International Conference on Machine Learning (ICML) 2014

• Topic Modeling from Network Spread
  A. Ray, S. Sanghavi, S. Shakkottai
  ACM SIGMETRICS 2014

• Serving content with unknown demand: the high-dimensional regime
  S. Moharir, S. Sanghavi, S. Shakkottai
  ACM SIGMETRICS 2014

2013

• Phase Retrieval using Alternating Minimization
  P. Netrapalli, P. Jain, S. Sanghavi
  Neural Information Processing Systems (NIPS) 2013

• Low-rank Matrix Completion using Alternating Minimization
  P. Netrapalli, P. Jain, S. Sanghavi
  Symposium on Theory of Computing (STOC) 2013

• Online load balancing under graph constraints
  S. Moharir, S. Sanghavi, S. Shakkottai
  ACM Sigmetrics 2013 (poster)

2012

• Clustering sparse graphs
  Y. Chen, S. Sanghavi, H. Xu
  Neural Information Processing Systems (NIPS) 2012

• Greedy learning of Graphical Models with Small Girth,
  A. Ray, S. Sanghavi, S. Shakkottai
  Allerton 2012.

• Online load balancing and correlated randomness,
  S. Moharir, S. Sanghavi
  Allerton 2012.

• Learning the Dependence Graph of Time Series with Latent Factors,
  A. Jalali, S. Sanghavi
  International Conference on Machine Learning (ICML), 2012.

• Learning the Graph of Epidemic Cascades
  P. Netrapalli, S. Sanghavi
  ACM SIGMETRICS, 2012.

• Low-rank Matrix Recovery from Errors and Erasures
  Y. Chen, A. Jalali, S. Sanghavi, C. Caramanis
  IEEE Transactions on Information Theory

• On the Effect of Channel Fading on Greedy Scheduling
  A. Reddy, S. Sanghavi, S. Shakkottai
  IEEE INFOCOM, 2012.

• Learning Markov Graphs Up To Edit Distance
  A. Das, P. Netrapalli, S. Sanghavi, S. Vishwanath
  IEEE International Symposium on Information Theory (ISIT), 2012.

2011

• Robust PCA via Outlier Pursuit
  H. Xu, C. Caramais, S. Sanghavi
  IEEE Transactions on Information Theory

• Low-rank Matrix Recovery from Errors and Erasures
  Y. Chen, A. Jalali, S. Sanghavi, C. Caramanis
  International Symposium on Information Theory (ISIT), 2011.

• Clustering Partially Observed Graphs via Convex Optimization
  A. Jalali, Y. Chen, S. Sanghavi, H. Xu
  International Conference on Machine Learning (ICML), 2011.

• Robust Matrix Completion with Corrupted Columns
  Y. Chen, H. Xu, C. Caramanis, S. Sanghavi
  International Conference on Machine Learning (ICML), 2011

• Rank-Sparsity Incoherence for Matrix Decomposition
  Venkat Chandrasekaran, Sujay Sanghavi, Pablo Parrilo, Alan S. Willsky
  SIAM Journal of Optimization, Vol. 21, No. 2, June 2011.

• Belief propagation and LP relaxation for weighted matching in general graphs
  S. Sanghavi, D. Malioutov, A. Willsky
  IEEE Transactions on Information Theory, vol. 57, iss. 4, pp 2203-12, 2011.

• On Learning Discrete Graphical Models using Group-Sparse Regularization
  A. Jalali, V. Vasuki, P. Ravikumar and S. Sanghavi
  Artifical Intelligence and Statistics (AISTATs) 2011

2010

• Robust PCA via Outlier Pursuit H. Xu, C. Caramais, S. Sanghavi
  Neural Information Processing Systems (NIPS), 2010

• A Dirty Model for Multitask Learning
  Ali Jalali, Pradeep Ravikumar, Sujay Sanghavi, Chao Ruan
  Neural Information Processing Systems (NIPS) 2010

• Learning Graphical Models for Hypothesis Testing and Classification
  V. Tan, S. Sanghavi, J. Fisher, A. Willsky
  IEEE Transactions on Signal Processing

• Greedy learning of Markov network structure
  P. Netrapalli, S. Banerjee, S. Sanghavi, S. Shakkottai
  Allerton Conference, 2010

• Sequential Compressed Sensing
  D. Malioutov, S. Sanghavi and A. Willsky
  IEEE Journal on Selected Areas in Signal Processing (special issue on Compressed Sensing)

2009

• Sparse and Low-Rank Matrix Decompositions
  Venkat Chandrasekaran, Sujay Sanghavi, Pablo Parrilo, Alan S. Willsky
  15th IFAC Sypmposium on System Identification (SYSID), 2009

• Is it enough to drain the heaviest bottlenecks?
  G. Gupta, S. Sanghavi, N. Shroff
  Allerton Conference, 2009

• Node Weighted Scheduling
  G. Gupta, S. Sanghavi, N. Shroff
  ACM SIGMETRICS, 2009

• Independent sets, Tightnessof LP and Max-product
  Sujay Sanghavi, Devavrat Shah
  accepted to AISTATS 2009 (withdrawn due to authors’ inability to travel to venue)

• Message Passing for Max-weight Independent Set
  S. Sanghavi, D. Shah, A. Willsky
  IEEE Transactions on Information Theory vol. 55, iss. 11, pp. 4822-34, 2009

• Distributed Link Scheduling with Constant Overhead
  L. Bui, S. Sanghavi, R. Srikant
  IEEE/ACM Transactions on Networking, vol. 17, iss. 5, pp 1467-80, 2009

2008

• Networking Sensors via Belief Propagation
  S. Sanghavi, D. Malioutov, A. Willsky
  Allerton Conference, 2008.

• Compressed Sensing with Sequential Observations
  D. Malioutov, S. Sanghavi, A. Willsky
  International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2008.

• A new mechanism for the free-rider problem
  S. Sanghavi, B. Hajek
  IEEE Transactions on Automatic Control, Vol. 53, No. 5, June 2008, pp 1176-1183.

2007

• Gossiping with Multiple Messages
  S. Sanghavi, B. Hajek and L. Massoulie
  IEEE Transactions on Information Theory, Vol. 53, Issue 12, Dec 2007, pp 4640-4654

• Linear Programming Analysis of Loopy Belief Propagation for Weighted Matching
  S. Sanghavi, D. Malioutov, A. Willsky
  Neural Information Processing Systems (NIPS), 2007

• Message Passing for Max-weight Independent Set
  S. Sanghavi, D. Shah and A. Willsky
  Neural Information Processing Systems (NIPS), 2007

• Intermediate Performance of Rateless Codes
  S. Sanghavi
  Information Theory Workshop (ITW), 2007.

• Equivalence of LP Relaxation and Max-Product for Weighted Matching in General Graphs
  S. Sanghavi
  Information Theory Workshop (ITW), 2007.

• Learning Graphical Models for Hypothesis Testing
  S. Sanghavi, V. Tan, A. Willsky
  Statistical Signal Processing Workshop (SSP), 2007

• Distributed Link Scheduling with Constant Overhead
  S. Sanghavi, L. Bui, R. Srikant
  ACM SIGMETRICS, 2007

• Gossiping with Multiple Messages
  S. Sanghavi, B. Hajek and L. Massoulie
  IEEE INFOCOM, 2007

2006 and before

• Adaptive Induced Fluctuations for Multiuser Diversity
  S. Sanghavi, B. Hajek
  IEEE Transactions on Wireless Communications

• A new mechanism for the free-rider problem
  S. Sanghavi, B. Hajek
  Third Sigcomm Workshop on Economics of Peer-to-Peer Systems (P2PECON), 2005

• Optimal Allocation of a Divisible Good to Strategic Buyers
  S. Sanghavi, B. Hajek
  IEEE Conference on Decision and Control (CDC), 2004

• Adaptive Induced Fluctuations for Multiuser Diversity
  S. Sanghavi, B. Hajek
  International Symposium on Information Theory (ISIT) 2002