High-dimensional linear regression via implicit regularization

Keyphrases

• High-dimensional Regression 100%
• Implicit Regularization 100%
• Gradient Descent 66%
• Signal-to-noise Ratio 33%
• System Dynamics 33%
• Discretized 33%
• Regularizer 33%
• Data Dependency 33%
• Rate-optimal 33%
• Statistical Estimators 33%
• M-estimator 33%
• Restricted Isometry 33%
• Class of Estimators 33%
• Residual Sum of Squares 33%
• Gradient Dynamics 33%
• Square Loss Function 33%
• Sparse Vector Estimation 33%
• Early Stopping Rules 33%
• Explicit Penalty 33%

Mathematics

• Regularization 100%
• Linear Regression 100%
• Parametric 25%
• Sum of Squares 25%
• Noise Ratio 25%
• Stopping Rule 25%
• Initial Value 25%
• Loss Function 25%
• M-Estimator 25%
• Dependent Data 25%
• Square Loss 25%
• Residual Sum 25%