What is he best known for?
The fields of study he is best known for:
- Statistics
- Algorithm
- Machine learning
Sujay Sanghavi mainly investigates Matrix, Algorithm, Combinatorics, Sparse matrix and Matrix completion.
Sujay Sanghavi studied Matrix and Convergence that intersect with Computational complexity theory and Low-rank approximation.
His Linear programming relaxation and Linear programming study in the realm of Algorithm connects with subjects such as Raptor code and Tornado code.
His Combinatorics research integrates issues from Stochastic block model, Clustering coefficient and Cluster analysis.
His research integrates issues of Matrix decomposition, Single-entry matrix, Discrete mathematics and Integer matrix in his study of Sparse matrix.
His Matrix completion study combines topics in areas such as Efficient algorithm, Collaborative filtering and Mathematical optimization, Minification.
His most cited work include:
- Rank-Sparsity Incoherence for Matrix Decomposition (818 citations)
- Low-rank matrix completion using alternating minimization (659 citations)
- Robust PCA via Outlier Pursuit (345 citations)
What are the main themes of his work throughout his whole career to date?
Algorithm, Combinatorics, Matrix, Mathematical optimization and Discrete mathematics are his primary areas of study.
Sujay Sanghavi combines subjects such as Linear regression and Outlier with his study of Algorithm.
The various areas that Sujay Sanghavi examines in his Combinatorics study include Stochastic block model, Clustering coefficient, Cluster analysis, Gradient descent and Upper and lower bounds.
As part of his studies on Matrix, Sujay Sanghavi frequently links adjacent subjects like Convex optimization.
Sujay Sanghavi interconnects Scheduling, Stochastic gradient descent and Throughput in the investigation of issues within Mathematical optimization.
His studies deal with areas such as Graphical model, Convex function, Markov chain, Applied mathematics and Relaxation as well as Discrete mathematics.
He most often published in these fields:
- Algorithm (28.09%)
- Combinatorics (24.72%)
- Matrix (23.60%)
What were the highlights of his more recent work (between 2017-2021)?
- Algorithm (28.09%)
- Upper and lower bounds (9.55%)
- Linear regression (7.87%)
In recent papers he was focusing on the following fields of study:
His primary areas of study are Algorithm, Upper and lower bounds, Linear regression, Combinatorics and Matrix.
His work on Compressed sensing as part of general Algorithm study is frequently connected to Fraction, therefore bridging the gap between diverse disciplines of science and establishing a new relationship between them.
His work carried out in the field of Linear regression brings together such families of science as Artificial neural network, Stochastic gradient descent, Artificial intelligence and Parameterized complexity.
His Parameterized complexity research incorporates themes from Minimum weight, Deep learning, Mathematical optimization and Conjecture.
His work in Combinatorics tackles topics such as Matrix decomposition which are related to areas like Rectified Gaussian distribution and Exponential function.
His Matrix research includes themes of Topic model and Moment.
Between 2017 and 2021, his most popular works were:
- Learning with Bad Training Data via Iterative Trimmed Loss Minimization (61 citations)
- Finding Low-Rank Solutions via Nonconvex Matrix Factorization, Efficiently and Provably (25 citations)
- Learning a Compressed Sensing Measurement Matrix via Gradient Unrolling (23 citations)
In his most recent research, the most cited papers focused on:
- Statistics
- Algorithm
- Machine learning
His primary areas of study are Algorithm, Matrix, Linear regression, Stochastic gradient descent and Minimum weight.
Sujay Sanghavi has included themes like Convergence and Quantum in his Algorithm study.
His Matrix research incorporates elements of Topic model and Combinatorics.
His Linear regression research integrates issues from Current, State, Gaussian and Linear model.
His Stochastic gradient descent research is multidisciplinary, incorporating elements of Parameterized complexity, Conjecture, Norm, Mathematical optimization and Generalization.
His Compressed sensing research includes elements of Embedding, Encoder, Subgradient method and Encoding.
This overview was generated by a machine learning system which analysed the scientist’s
body of work. If you have any feedback, you can
contact us
here.
