ROOT-SGD: Sharp Nonasymptotics and Asymptotic Efficiency in a Single Algorithm.
Chris Junchi LiWenlong MouMartin J. WainwrightMichael I. JordanPublished in: COLT (2022)
Keyphrases
- learning algorithm
- computational complexity
- worst case
- detection algorithm
- preprocessing
- cost function
- experimental evaluation
- optimal solution
- high efficiency
- recognition algorithm
- significant improvement
- times faster
- matching algorithm
- computationally efficient
- linear programming
- objective function
- computational efficiency
- high accuracy
- dynamic programming
- search space
- simulated annealing
- expectation maximization
- similarity measure
- monte carlo
- neural network
- probabilistic model
- k means
- highly efficient