Výsledky vyhľadávania - "stochastic variance-reduced algorithm"

Faster First-Order Methods for Stochastic Non-Convex Optimization on Riemannian Manifolds Autor Zhou, Pan, Yuan, Xiao-Tong, Yan, Shuicheng, Feng, Jiashi

“…First-order non-convex Riemannian optimization algorithms have gained recent popularity in structured machine learning problems including principal component…”
Získať plný text

A Hybrid Stochastic-Deterministic Minibatch Proximal Gradient Method for Efficient Optimization and Generalization Autor Zhou, Pan, Yuan, Xiao-Tong, Lin, Zhouchen, Hoi, Steven C.H.

“…Despite the success of stochastic variance-reduced gradient (SVRG) algorithms in solving large-scale problems, their stochastic gradient complexity often…”
Získať plný text

A Simple Stochastic Variance Reduced Algorithm with Fast Convergence Rates Autor Zhou, Kaiwen, Shang, Fanhua, Cheng, James

“…, sparse and asynchronous) due to the existence of perturbation. In this paper, we introduce a simple stochastic variance reduced algorithm (MiG…”
Získať plný text

Accelerated variance-reduced methods for saddle-point problems Autor Borodich, Ekaterina, Tominin, Vladislav, Tominin, Yaroslav, Kovalev, Dmitry, Gasnikov, Alexander, Dvurechensky, Pavel

“… for the primal and dual variables. For such problems, under the average-smoothness assumption, we propose accelerated stochastic variance-reduced algorithms with optimal up to logarithmic factors complexity bounds…”
Získať plný text

Byzantine Resilient Non-Convex SCSG With Distributed Batch Gradient Computations Autor Bulusu, Saikiran, Khanduri, Prashant, Kafle, Swatantra, Sharma, Pranay, Varshney, Pramod K.

“… A robust variant of the stochastic variance-reduced algorithm is proposed. In the distributed setup, we assume that a fraction of worker nodes (WNs) can be Byzantines…”
Získať plný text

Stochastic Scale Invariant Power Iteration for KL-divergence Nonnegative Matrix Factorization Autor Kim, Cheolmin, Kim, Youngseok, Jambunath, Yegna Subramanian, Klabjan, Diego

“…We introduce a mini-batch stochastic variance-reduced algorithm to solve finite-sum scale invariant problems which cover several examples in machine learning and statistics such as principal component analysis (PCA…”
Získať plný text

Distributionally Adversarial Learning Autor Shi, Zhan

“… Then we develop a new stochastic variance-reduced algorithm to efficiently solve them, which allows any Bregman divergence as a proximal function and achieves linear convergence rates…”
Získať plný text

Variance-reduction for Variational Inequality Problems with Bregman Distance Function Autor Alizadeh, Zeinab, Erfan Yazdandoost Hamedani, Jalilzadeh, Afrooz

“… We introduce a novel single-loop stochastic variance-reduced algorithm, incorporating the Bregman distance function, and establish an optimal convergence guarantee under a monotone setting…”
Získať plný text

Stochastic Scale Invariant Power Iteration for KL-divergence Nonnegative Matrix Factorization Autor Kim, Cheolmin, Kim, Youngseok, Klabjan, Diego

“…We introduce a mini-batch stochastic variance-reduced algorithm to solve finite-sum scale invariant problems which cover several examples in machine learning and statistics such as principal component analysis (PCA…”
Získať plný text

Stochastic variance reduced gradient with hyper-gradient for non-convex large-scale learning Autor Yang, Zhuang

“… With faster convergence rate, there have been tremendous studies on developing stochastic variance reduced algorithms to solve these non-convex optimization problems…”
Získať plný text

Convergence of Distributed Stochastic Variance Reduced Methods Without Sampling Extra Data Autor Cen, Shicong, Zhang, Huishuai, Chi, Yuejie, Chen, Wei, Liu, Tie-Yan

“…Stochastic variance reduced methods have gained a lot of interest recently for empirical risk minimization due to its appealing run time complexity. When the…”
Získať plný text

Convergence of Distributed Stochastic Variance Reduced Methods without Sampling Extra Data Autor Cen, Shicong, Zhang, Huishuai, Chi, Yuejie, Chen, Wei, Tie-Yan, Liu

“…Stochastic variance reduced methods have gained a lot of interest recently for empirical risk minimization due to its appealing run time complexity. When the…”
Získať plný text

Complexity Theory of Stochastic Algorithms for Large-Scale Nonconvex Optimization Autor Wang, Zhe

“… For the first-order algorithms, SARAH and SPIDER are two recently developed stochastic variance reduced algorithms, and SPIDER has been shown to achieve a near-optimal first-order oracle complexity…”
Získať plný text

Momentum Schemes with Stochastic Variance Reduction for Nonconvex Composite Optimization Autor Zhou, Yi, Wang, Zhe, Ji, Kaiyi, Liang, Yingbin, Tarokh, Vahid

“…Two new stochastic variance-reduced algorithms named SARAH and SPIDER have been recently proposed, and SPIDER has been shown to achieve a near-optimal gradient oracle complexity for nonconvex optimization…”
Získať plný text

Stochastic Variance Reduction for Variational Inequality Methods Autor Alacaoglu, Ahmet, Malitsky, Yura

“…We propose stochastic variance reduced algorithms for solving convex-concave saddle point problems, monotone variational inequalities, and monotone inclusions…”
Získať plný text

An Accelerated Variance Reduced Extra-Point Approach to Finite-Sum VI and Optimization Autor Huang, Kevin, Wang, Nuozhou, Zhang, Shuzhong

“…In this paper, we develop stochastic variance reduced algorithms for solving a class of finite-sum monotone VI, where the operator consists of the sum of finitely many monotone VI mappings and the sum…”
Získať plný text

Dual-Free Stochastic Decentralized Optimization with Variance Reduction Autor Hendrikx, Hadrien, Bach, Francis, Massoulié, Laurent

“… DVR only requires computing stochastic gradients of the local functions, and is computationally as fast as a standard stochastic variance-reduced algorithms run on a \(1/n…”
Získať plný text

SpiderBoost and Momentum: Faster Stochastic Variance Reduction Algorithms Autor Wang, Zhe, Ji, Kaiyi, Zhou, Yi, Liang, Yingbin, Tarokh, Vahid

“…SARAH and SPIDER are two recently developed stochastic variance-reduced algorithms, and SPIDER has been shown to achieve a near-optimal first-order oracle complexity in smooth nonconvex optimization…”
Získať plný text