Learning rate adaptive stochastic gradient descent optimization methods: numerical simulations for deep learning methods for partial differential equations and convergence analyses.
Steffen DereichArnulf JentzenAdrian RiekertPublished in: CoRR (2024)
Keyphrases
- optimization methods
- numerical simulations
- learning rate
- partial differential equations
- global convergence
- deep learning
- stochastic gradient descent
- convergence rate
- convergence speed
- simulated annealing
- optimization problems
- optimization method
- step size
- learning algorithm
- theoretical analysis
- loss function
- online learning
- level set
- denoising
- multi class
- support vector machine
- pattern recognition