stochastic gradient

A Near-Optimal Algorithm for Stochastic Bilevel Optimization via Double-Momentum NeurIPS 2021 A Scalable Gradient Free Method for Bayesian Experimental Design with Implicit Models AISTATS 2021 A Stochastic Newton Algorithm for Distributed Convex Optimization NeurIPS 2021 Decentralized Stochastic Gradient Langevin Dynamics and Hamiltonian Monte Carlo JMLR 2021 Differentiable Annealed Importance Sampling and the Perils of Gradient Noise NeurIPS 2021 Distributed Saddle-Point Problems Under Data Similarity NeurIPS 2021 Dynamics of Stochastic Momentum Methods on Large-scale, Quadratic Models NeurIPS 2021 EF21: A New, Simpler, Theoretically Better, and Practically Faster Error Feedback NeurIPS 2021 Efficient Mirror Descent Ascent Methods for Nonsmooth Minimax Problems NeurIPS 2021 Federated Learning with Compression: Unified Analysis and Sharp Guarantees AISTATS 2021 Federated stochastic gradient Langevin dynamics UAI 2021 Finding First-Order Nash Equilibria of Zero-Sum Games with the Regularized Nikaido-Isoda Function AISTATS 2021 Fractional moment-preserving initialization schemes for training deep neural networks AISTATS 2021 High-Order Langevin Diffusion Yields an Accelerated MCMC Algorithm JMLR 2021 JEM++: Improved Techniques for Training JEM ICCV 2021 Langevin Dynamics for Adaptive Inverse Reinforcement Learning of Stochastic Gradient Algorithms JMLR 2021 Langevin Monte Carlo: random coordinate descent and variance reduction JMLR 2021 Learning from Censored and Dependent Data: The case of Linear Dynamics COLT 2021 Learning with risk-averse feedback under potentially heavy tails AISTATS 2021 MetaGrad: Adaptation using Multiple Learning Rates in Online Learning JMLR 2021 Near Optimal Policy Optimization via REPS NeurIPS 2021 On the Bias-Variance-Cost Tradeoff of Stochastic Optimization NeurIPS 2021 On the Convergence and Sample Efficiency of Variance-Reduced Policy Gradient Method NeurIPS 2021 On the Convergence of Hamiltonian Monte Carlo with Stochastic Gradients ICML 2021 On the Convergence Theory of Debiased Model-Agnostic Meta-Reinforcement Learning NeurIPS 2021