Improving sampling accuracy of stochastic gradient MCMC methods via non-uniform subsampling of gradients
DOI10.3934/dcdss.2021157zbMath1514.65008arXiv2002.08949OpenAlexW3007353463MaRDI QIDQ6160667
Molei Tao, Hongyuan Zha, Xin Wang, Ruilin Li
Publication date: 26 June 2023
Published in: Discrete and Continuous Dynamical Systems. Series S (Search for Journal in Brave)
Full work available at URL: https://arxiv.org/abs/2002.08949
bias-variance trade-offMCMC sampling and Bayesian inferencenon-uniform weighted stochastic gradientnonasymptotic sampling error boundtransition kernel matching
Sampling theory, sample surveys (62D05) Monte Carlo methods (65C05) Stochastic programming (90C15) Applications of Markov chains and discrete-time Markov processes on general state spaces (social mobility, learning theory, industrial processes, etc.) (60J20) Numerical analysis or methods applied to Markov chains (65C40) Dynamical systems and their relations with probability theory and stochastic processes (37A50) Dynamical aspects of statistical mechanics (37A60)
Cites Work
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Minimizing finite sums with the stochastic average gradient
- A strong approximation theorem for stochastic recursive algorithms
- Exponential convergence of Langevin distributions and their discrete approximations
- User-friendly guarantees for the Langevin Monte Carlo with inaccurate gradient
- Control variates for stochastic gradient MCMC
- Coupling and convergence for Hamiltonian Monte Carlo
- (Non-) asymptotic properties of Stochastic Gradient Langevin Dynamics
- Long-Run Accuracy of Variational Integrators in the Stochastic Context
- Convergence of Numerical Time-Averaging and Stationary Measures via Poisson Equations
- Handbook of Markov Chain Monte Carlo
- Stochastic Processes and Applications
- Geometric integrators and the Hamiltonian Monte Carlo method
- Stochastic gradient descent, weighted sampling, and the randomized Kaczmarz algorithm
