Accelerating Convergence of Replica Exchange Stochastic Gradient MCMC via Variance Reduction
(2021)
Presentation / Conference Contribution
Deng, W., Feng, Q., Karagiannis, G., Lin, G., & Liang, F. (2021, December). Accelerating Convergence of Replica Exchange Stochastic Gradient MCMC via Variance Reduction. Paper presented at International Conference on Learning Representations (ICLR'21), Virtual Event
Replica exchange stochastic gradient Langevin dynamics (reSGLD) has shown promise in accelerating the convergence in non-convex learning; however, an excessively large correction for avoiding biases from noisy energy estimators has limited the potent... Read More about Accelerating Convergence of Replica Exchange Stochastic Gradient MCMC via Variance Reduction.