On the Proof of Global Convergence of Gradient Descent for Deep ReLU Networks with Linear Widths.
Quynh NguyenPublished in: CoRR (2021)
Keyphrases
- global convergence
- global optimum
- convergence analysis
- convergence rate
- optimization methods
- convergence speed
- constrained optimization problems
- convex minimization
- objective function
- cost function
- line search
- social networks
- newton method
- hybrid algorithm
- network structure
- globally convergent
- global optimization
- least squares
- particle swarm
- conjugate gradient
- coordinate ascent
- machine learning