Corpus ID: 222080194

Variance reduction for Random Coordinate Descent-Langevin Monte Carlo.

  title={Variance reduction for Random Coordinate Descent-Langevin Monte Carlo.},
  author={Zhiyan Ding and Q. Li},
  journal={arXiv: Machine Learning},
  • Zhiyan Ding, Q. Li
  • Published 2020
  • Mathematics, Computer Science
  • arXiv: Machine Learning
  • Sampling from a log-concave distribution function is one core problem that has wide applications in Bayesian statistics and machine learning. While most gradient free methods have slow convergence rate, the Langevin Monte Carlo (LMC) that provides fast convergence requires the computation of gradients. In practice one uses finite-differencing approximations as surrogates, and the method is expensive in high-dimensions. A natural strategy to reduce computational cost in each iteration is to… CONTINUE READING


    Publications referenced by this paper.
    Monte Carlo Sampling Methods Using Markov Chains and Their Applications
    • 11,923
    • PDF
    Accelerating Stochastic Gradient Descent using Predictive Variance Reduction
    • 1,535
    • PDF
    Stochastic Gradient Hamiltonian Monte Carlo
    • 431
    • Highly Influential
    • PDF
    Sequential Monte Carlo Methods in Practice
    • 3,525
    Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
    • 11,727
    • PDF
    Annealed importance sampling
    • R. Neal
    • Mathematics, Computer Science
    • 2001
    • 1,049
    • PDF
    Variance Reduction in Stochastic Gradient Langevin Dynamics
    • 52
    • PDF
    Langevin Monte Carlo: random coordinate descent and variance reduction
    • 1
    • PDF
    A Complete Recipe for Stochastic Gradient MCMC
    • 233
    • PDF
    User-friendly guarantees for the Langevin Monte Carlo with inaccurate gradient
    • 118
    • Highly Influential
    • PDF