Alert button
Picture for Michael W. Mahoney

Michael W. Mahoney

Alert button

Bayesian experimental design using regularized determinantal point processes

Jun 10, 2019
Michał Dereziński, Feynman Liang, Michael W. Mahoney

Figure 1 for Bayesian experimental design using regularized determinantal point processes
Figure 2 for Bayesian experimental design using regularized determinantal point processes
Figure 3 for Bayesian experimental design using regularized determinantal point processes
Figure 4 for Bayesian experimental design using regularized determinantal point processes
Viaarxiv icon

Residual Networks as Nonlinear Systems: Stability Analysis using Linearization

May 31, 2019
Kai Rothauge, Zhewei Yao, Zixi Hu, Michael W. Mahoney

Figure 1 for Residual Networks as Nonlinear Systems: Stability Analysis using Linearization
Figure 2 for Residual Networks as Nonlinear Systems: Stability Analysis using Linearization
Figure 3 for Residual Networks as Nonlinear Systems: Stability Analysis using Linearization
Figure 4 for Residual Networks as Nonlinear Systems: Stability Analysis using Linearization
Viaarxiv icon

Distributed estimation of the inverse Hessian by determinantal averaging

May 28, 2019
Michał Dereziński, Michael W. Mahoney

Figure 1 for Distributed estimation of the inverse Hessian by determinantal averaging
Figure 2 for Distributed estimation of the inverse Hessian by determinantal averaging
Viaarxiv icon

Physics-informed Autoencoders for Lyapunov-stable Fluid Flow Prediction

May 26, 2019
N. Benjamin Erichson, Michael Muehlebach, Michael W. Mahoney

Figure 1 for Physics-informed Autoencoders for Lyapunov-stable Fluid Flow Prediction
Figure 2 for Physics-informed Autoencoders for Lyapunov-stable Fluid Flow Prediction
Figure 3 for Physics-informed Autoencoders for Lyapunov-stable Fluid Flow Prediction
Figure 4 for Physics-informed Autoencoders for Lyapunov-stable Fluid Flow Prediction
Viaarxiv icon

JumpReLU: A Retrofit Defense Strategy for Adversarial Attacks

Apr 07, 2019
N. Benjamin Erichson, Zhewei Yao, Michael W. Mahoney

Figure 1 for JumpReLU: A Retrofit Defense Strategy for Adversarial Attacks
Figure 2 for JumpReLU: A Retrofit Defense Strategy for Adversarial Attacks
Figure 3 for JumpReLU: A Retrofit Defense Strategy for Adversarial Attacks
Figure 4 for JumpReLU: A Retrofit Defense Strategy for Adversarial Attacks
Viaarxiv icon

OverSketched Newton: Fast Convex Optimization for Serverless Systems

Mar 21, 2019
Vipul Gupta, Swanand Kadhe, Thomas Courtade, Michael W. Mahoney, Kannan Ramchandran

Figure 1 for OverSketched Newton: Fast Convex Optimization for Serverless Systems
Figure 2 for OverSketched Newton: Fast Convex Optimization for Serverless Systems
Figure 3 for OverSketched Newton: Fast Convex Optimization for Serverless Systems
Figure 4 for OverSketched Newton: Fast Convex Optimization for Serverless Systems
Viaarxiv icon

Inefficiency of K-FAC for Large Batch Size Training

Mar 14, 2019
Linjian Ma, Gabe Montague, Jiayu Ye, Zhewei Yao, Amir Gholami, Kurt Keutzer, Michael W. Mahoney

Figure 1 for Inefficiency of K-FAC for Large Batch Size Training
Figure 2 for Inefficiency of K-FAC for Large Batch Size Training
Figure 3 for Inefficiency of K-FAC for Large Batch Size Training
Figure 4 for Inefficiency of K-FAC for Large Batch Size Training
Viaarxiv icon

Shallow Learning for Fluid Flow Reconstruction with Limited Sensors and Limited Data

Feb 20, 2019
N. Benjamin Erichson, Lionel Mathelin, Zhewei Yao, Steven L. Brunton, Michael W. Mahoney, J. Nathan Kutz

Figure 1 for Shallow Learning for Fluid Flow Reconstruction with Limited Sensors and Limited Data
Figure 2 for Shallow Learning for Fluid Flow Reconstruction with Limited Sensors and Limited Data
Figure 3 for Shallow Learning for Fluid Flow Reconstruction with Limited Sensors and Limited Data
Figure 4 for Shallow Learning for Fluid Flow Reconstruction with Limited Sensors and Limited Data
Viaarxiv icon

Minimax experimental design: Bridging the gap between statistical and worst-case approaches to least squares regression

Feb 04, 2019
Michał Dereziński, Kenneth L. Clarkson, Michael W. Mahoney, Manfred K. Warmuth

Viaarxiv icon

Heavy-Tailed Universality Predicts Trends in Test Accuracies for Very Large Pre-Trained Deep Neural Networks

Jan 24, 2019
Charles H. Martin, Michael W. Mahoney

Figure 1 for Heavy-Tailed Universality Predicts Trends in Test Accuracies for Very Large Pre-Trained Deep Neural Networks
Figure 2 for Heavy-Tailed Universality Predicts Trends in Test Accuracies for Very Large Pre-Trained Deep Neural Networks
Figure 3 for Heavy-Tailed Universality Predicts Trends in Test Accuracies for Very Large Pre-Trained Deep Neural Networks
Figure 4 for Heavy-Tailed Universality Predicts Trends in Test Accuracies for Very Large Pre-Trained Deep Neural Networks
Viaarxiv icon