Global convergence of non-convex gradient descent for computing matrix squareroot

Keyphrases

• Global Convergence 100%
• Nonconvex 100%
• Gradient Descent 100%
• Machine Learning 50%
• Gradient Method 50%
• Positive Definite Matrix 50%
• Numerical Linear Algebra 50%
• Convergence Rate 25%
• Proof Techniques 25%
• Positive Definite 25%
• Nonconvex Optimization 25%
• Amount of Work 25%
• Take-the-first 25%
• Local Convergence 25%
• Global Convergence Rate 25%
• Rate Convergence 25%

Mathematics

• Matrix (Mathematics) 100%
• Positive Definite Matrix 100%
• Numerical Linear Algebra 100%
• Stochastics 50%
• Statistics 50%
• Positive Definite 50%
• Starting Point 50%
• Convergence Rate 50%
• Local Convergence 50%