Maya R. Gupta
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Maya R. Gupta has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
Erika Chin
3 shared publications- Luca Cazzanti
Dept. of Electrical Engineering University of Washington Seattle 98195 WA USA
3 shared publications - Santosh Srivastava
Department of Applied Mathematics, University of Washington, Seattle, WA 98195, USA
3 shared publications - Eric K. Garcia
Department of Electrical Engineering, University of Washington, Seattle, WA 98195, USA
2 shared publications - Jill K. Nelson
George Mason University
2 shared publications Andre B. Maachar
1 shared publication
Latest Publications
- Open Access
Dimensionality Reduction by Local Discriminative Gaussians
We present local discriminative Gaussian (LDG) dimensionality reduction, a supervised dimensionality reduction technique for classification. The LDG objective function is an approximation to the leave-one-out training error of a local quadratic discriminant analysis classifier, and thus acts locally to each training point in order to find a mapping where similar data can be discriminated from dissimilar data. While other state-of-the-art linear dimensionality reduction methods require gradient descent or iterative solution approaches, LDG is solved with a single eigen-decomposition. Thus, it scales better for datasets with a large number of feature dimensions or training examples. We also adapt LDG to the transfer learning setting, and show that it achieves good performance when the test data distribution differs from that of the training data.
Open AccessDimensionality Reduction by Local Discriminative Gaussians
We present local discriminative Gaussian (LDG) dimensionality reduction, a supervised dimensionality reduction technique for classification. The LDG objective function is an approximation to the leave-one-out training error of a local quadratic discriminant analysis classifier, and thus acts locally to each training point in order to find a mapping where similar data can be discriminated from dissimilar data. While other state-of-the-art linear dimensionality reduction methods require gradient descent or iterative solution approaches, LDG is solved with a single eigen-decomposition. Thus, it scales better for datasets with a large number of feature dimensions or training examples. We also adapt LDG to the transfer learning setting, and show that it achieves good performance when the test data distribution differs from that of the training data.
- Open Access
Bounds on the Bayes Error Given Moments
We show how to compute lower bounds for the supremum Bayes error if the class-conditional distributions must satisfy moment constraints, where the supremum is with respect to the unknown class-conditional distributions. Our approach makes use of Curto and Fialkow's solutions for the truncated moment problem. The lower bound shows that the popular Gaussian assumption is not robust in this regard. We also construct an upper bound for the supremum Bayes error by constraining the decision boundary to be linear.
Open AccessBounds on the Bayes Error Given Moments
We show how to compute lower bounds for the supremum Bayes error if the class-conditional distributions must satisfy moment constraints, where the supremum is with respect to the unknown class-conditional distributions. Our approach makes use of Curto and Fialkow's solutions for the truncated moment problem. The lower bound shows that the popular Gaussian assumption is not robust in this regard. We also construct an upper bound for the supremum Bayes error by constraining the decision boundary to be linear.
- Top Cited Papers
- 15 April 2011
- Published by Emerald Publishing in Foundations and Trends® in Signal Processing
- Vol. 4 (3) , 223-296
Theory and Use of the EM Algorithm
This introduction to the expectation–maximization (EM) algorithm provides an intuitive and mathematically rigorous understanding of EM. Two of the most popular applications of EM are described in detail: estimating Gaussian mixture models (GMMs), and estimating hidden Markov models (HMMs). EM solutions are also derived for learning an optimal mixture of fixed models, for estimating the parameters of a compound Dirichlet distribution, and for dis-entangling superimposed signals. Practical issues that arise in the use of EM are discussed, as well as variants of the algorithm that help deal with these challenges.
- 15 April 2011
- Published by Emerald Publishing in Foundations and Trends® in Signal Processing
- Vol. 4 (3) , 223-296
Top Cited PapersTheory and Use of the EM Algorithm
This introduction to the expectation–maximization (EM) algorithm provides an intuitive and mathematically rigorous understanding of EM. Two of the most popular applications of EM are described in detail: estimating Gaussian mixture models (GMMs), and estimating hidden Markov models (HMMs). EM solutions are also derived for learning an optimal mixture of fixed models, for estimating the parameters of a compound Dirichlet distribution, and for dis-entangling superimposed signals. Practical issues that arise in the use of EM are discussed, as well as variants of the algorithm that help deal with these challenges.
- 16 March 2011
- Published by Springer Nature
- p. 65-79
Color Science and Engineering for the Display of Remote Sensing Images
- 16 March 2011
- Published by Springer Nature
- p. 65-79
Color Science and Engineering for the Display of Remote Sensing Images
- Open Access
Parametric Bayesian Estimation of Differential Entropy and Relative Entropy
Given iid samples drawn from a distribution with known parametric form, we propose the minimization of expected Bregman divergence to form Bayesian estimates of differential entropy and relative entropy, and derive such estimators for the uniform, Gaussian, Wishart, and inverse Wishart distributions. Additionally, formulas are given for a log gamma Bregman divergence and the differential entropy and relative entropy for the Wishart and inverse Wishart. The results, as always with Bayesian estimates, depend on the accuracy of the prior parameters, but example simulations show that the performance can be substantially improved compared to maximum likelihood or state-of-the-art nonparametric estimators.
Open AccessParametric Bayesian Estimation of Differential Entropy and Relative Entropy
Given iid samples drawn from a distribution with known parametric form, we propose the minimization of expected Bregman divergence to form Bayesian estimates of differential entropy and relative entropy, and derive such estimators for the uniform, Gaussian, Wishart, and inverse Wishart distributions. Additionally, formulas are given for a log gamma Bregman divergence and the differential entropy and relative entropy for the Wishart and inverse Wishart. The results, as always with Bayesian estimates, depend on the accuracy of the prior parameters, but example simulations show that the performance can be substantially improved compared to maximum likelihood or state-of-the-art nonparametric estimators.
- 30 May 2008
- Published by Optica Publishing Group in Journal of the Optical Society of America A
- Vol. 25 (6) , 1454-1458
Ranked dither for high-quality robust printing.
A robust parallelized halftoning method is proposed that is insensitive to dot loss, which can be a problem with xerographic and high-resolution printing. The method uses standard clustered-dot dither for smooth image blocks but uses the rankings of dither thresholds to halftone image blocks with high spatial activity. Experiments compare the ranked dither to dither, error diffusion, and green-noise error diffusion for four-color printing using standard clustered-dot dither masks at standard angles for the cyan, magenta, yellow, and black planes, respectively.
- 30 May 2008
- Published by Optica Publishing Group in Journal of the Optical Society of America A
- Vol. 25 (6) , 1454-1458
Ranked dither for high-quality robust printing.
A robust parallelized halftoning method is proposed that is insensitive to dot loss, which can be a problem with xerographic and high-resolution printing. The method uses standard clustered-dot dither for smooth image blocks but uses the rankings of dither thresholds to halftone image blocks with high spatial activity. Experiments compare the ranked dither to dither, error diffusion, and green-noise error diffusion for four-color printing using standard clustered-dot dither masks at standard angles for the cyan, magenta, yellow, and black planes, respectively.
- 12 May 2008
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 17 (6) , 936-945
Adaptive Local Linear Regression With Application to Printer Color Management
Local learning methods, such as local linear regression and nearest neighbor classifiers, base estimates on nearby training samples, neighbors. Usually, the number of neighbors used in estimation is fixed to be a global ldquooptimalrdquo value, chosen by cross validation. This paper proposes adapting the number of neighbors used for estimation to the local geometry of the data, without need for cross validation. The term enclosing neighborhood is introduced to describe a set of neighbors whose convex hull contains the test point when possible. It is proven that enclosing neighborhoods yield bounded estimation variance under some assumptions. Three such enclosing neighborhood definitions are presented: natural neighbors, natural neighbors inclusive, and enclosing k-NN. The effectiveness of these neighborhood definitions with local linear regression is tested for estimating lookup tables for color management. Significant improvements in error metrics are shown, indicating that enclosing neighborhoods may be a promising adaptive neighborhood definition for other local learning tasks as well, depending on the density of training samples.
- 12 May 2008
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 17 (6) , 936-945
Adaptive Local Linear Regression With Application to Printer Color Management
Local learning methods, such as local linear regression and nearest neighbor classifiers, base estimates on nearby training samples, neighbors. Usually, the number of neighbors used in estimation is fixed to be a global ldquooptimalrdquo value, chosen by cross validation. This paper proposes adapting the number of neighbors used for estimation to the local geometry of the data, without need for cross validation. The term enclosing neighborhood is introduced to describe a set of neighbors whose convex hull contains the test point when possible. It is proven that enclosing neighborhoods yield bounded estimation variance under some assumptions. Three such enclosing neighborhood definitions are presented: natural neighbors, natural neighbors inclusive, and enclosing k-NN. The effectiveness of these neighborhood definitions with local linear regression is tested for estimating lookup tables for color management. Significant improvements in error metrics are shown, indicating that enclosing neighborhoods may be a promising adaptive neighborhood definition for other local learning tasks as well, depending on the density of training samples.
- Open Access
- 31 March 2008
- Published by Defense Technical Information Center (DTIC)
Lebanon: Strategic and Political Challenges and Opportunities
Local learning methods, such as local linear regression and nearest neighbor classifiers, base estimates on nearby training samples, neighbors. Usually the number of neighbors used in estimation is fixed to be a global optimal value, chosen by cross-validation. This paper proposes adapting the number of neighbors used for estimation to the local geometry of the data, without need for cross-validation. The term enclosing neighborhood is introduced to describe a set of neighbors whose convex hull contains the test point when possible. It is proven that enclosing neighborhoods yield bounded estimation variance under some assumptions. Three such enclosing neighborhood definitions are presented: natural neighbors, natural neighbors inclusive, and enclosing kappa-NN. The effectiveness of these neighborhood definitions with local linear regression is tested for estimating look-up tables for color management. Significant improvements in error metrics are shown, indicating that enclosing neighborhoods may be a promising adaptive neighborhood definition for other local learning tasks as well, depending on the density of training samples.
- 31 March 2008
- Published by Defense Technical Information Center (DTIC)
Open AccessLebanon: Strategic and Political Challenges and Opportunities
Local learning methods, such as local linear regression and nearest neighbor classifiers, base estimates on nearby training samples, neighbors. Usually the number of neighbors used in estimation is fixed to be a global optimal value, chosen by cross-validation. This paper proposes adapting the number of neighbors used for estimation to the local geometry of the data, without need for cross-validation. The term enclosing neighborhood is introduced to describe a set of neighbors whose convex hull contains the test point when possible. It is proven that enclosing neighborhoods yield bounded estimation variance under some assumptions. Three such enclosing neighborhood definitions are presented: natural neighbors, natural neighbors inclusive, and enclosing kappa-NN. The effectiveness of these neighborhood definitions with local linear regression is tested for estimating look-up tables for color management. Significant improvements in error metrics are shown, indicating that enclosing neighborhoods may be a promising adaptive neighborhood definition for other local learning tasks as well, depending on the density of training samples.
- Open Access
- 25 March 2008
- Published by Defense Technical Information Center (DTIC)
Command and Control Organizations for Network Enabled Warfare
We consider the problems of class probability estimation and classification when using near-neighbor classifiers, such as k-nearest neighbors (kNN). This paper investigates minimum expected risk estimates for neighborhood learning methods. We give analytic solutions for the minimum expected risk estimate for weighted kNN classifiers with different prior information, for a broad class of risk functions. Theory and simulations show how significant the difference is compared to the standard maximum likelihood weighted kNN estimates. Comparisons are made with uniform weights, symmetric weights (tricube kernel), and asymmetric weights (LIME kernel). Also, it is shown that if the uncertainty in the class probability is modeled by a random variable, and the expected misclassification cost is minimized, the result is equivalent to using a classifier with a minimum expected risk estimate. For symmetric costs and uniform priors, it is seen that minimum expected risk estimates have no advantage over the standard maximum likelihood estimates. For asymmetric costs, simulations show that the differences can be striking.
- 25 March 2008
- Published by Defense Technical Information Center (DTIC)
Open AccessCommand and Control Organizations for Network Enabled Warfare
We consider the problems of class probability estimation and classification when using near-neighbor classifiers, such as k-nearest neighbors (kNN). This paper investigates minimum expected risk estimates for neighborhood learning methods. We give analytic solutions for the minimum expected risk estimate for weighted kNN classifiers with different prior information, for a broad class of risk functions. Theory and simulations show how significant the difference is compared to the standard maximum likelihood weighted kNN estimates. Comparisons are made with uniform weights, symmetric weights (tricube kernel), and asymmetric weights (LIME kernel). Also, it is shown that if the uncertainty in the class probability is modeled by a random variable, and the expected misclassification cost is minimized, the result is equivalent to using a classifier with a minimum expected risk estimate. For symmetric costs and uniform priors, it is seen that minimum expected risk estimates have no advantage over the standard maximum likelihood estimates. For asymmetric costs, simulations show that the differences can be striking.
- Open Access
- 1 January 2008
- Published by Defense Technical Information Center (DTIC)
Adaptive Local Linear Regression with Application to Printer Color Management
This project assessed the biomedical and toxicological literature for evidence of hormesis, its frequency in the literature and its underlying mechanistic foundation. This work was supported by the continued development of the hormesis database and the conduct of a high level international conference on hormesis held annually. Particular focus was given to the area of neuroscience and hormesis in the literature assessment. Fourteen manuscripts concerning hormesis and neuroscience have been accepted for publication in the journal Critical Reviews in Toxicology and will be published in 2008.
- 1 January 2008
- Published by Defense Technical Information Center (DTIC)
Open AccessAdaptive Local Linear Regression with Application to Printer Color Management
This project assessed the biomedical and toxicological literature for evidence of hormesis, its frequency in the literature and its underlying mechanistic foundation. This work was supported by the continued development of the hormesis database and the conduct of a high level international conference on hormesis held annually. Particular focus was given to the area of neuroscience and hormesis in the literature assessment. Fourteen manuscripts concerning hormesis and neuroscience have been accepted for publication in the journal Critical Reviews in Toxicology and will be published in 2008.