A Simple Proof of the Mixing of Metropolis-Adjusted Langevin Algorithm under Smoothness and Isoperimetry.
Yuansi ChenKhashayar GatmiryPublished in: CoRR (2023)
Keyphrases
- cost function
- learning algorithm
- objective function
- estimation algorithm
- expectation maximization
- np hard
- theoretical analysis
- em algorithm
- matching algorithm
- times faster
- high accuracy
- worst case
- improved algorithm
- preprocessing
- probabilistic model
- k means
- detection algorithm
- prior information
- recognition algorithm
- classification algorithm
- provably correct
- parameter estimation
- computationally efficient
- particle swarm optimization
- input data
- moving objects
- optimal solution
- optimization algorithm
- monte carlo
- linear programming
- computational cost
- experimental evaluation
- neural network