paper-with-me

홈 › Papers

Averaging Stochastic Gradient Descent on Riemannian Manifolds

2018-02-26 · Nilesh Tripuraneni, Nicolas Flammarion, Francis Bach, Michael. I. Jordan

We consider the minimization of a function defined on a Riemannian manifold $\mathcal{M}$ accessible only through unbiased estimates of its gradients. We develop a geometric framework to transform a sequence of slowly converging iterates generated from stochastic gradient descent (SGD) on $\mathcal{M}$ to an averaged iterate sequence with a robust and fast $O(1/n)$ convergence rate. We then present an application of our framework to geodesically-strongly-convex (and possibly Euclidean non-convex) problems. Finally, we demonstrate how these ideas apply to the case of streaming $k$-PCA, where we show how to accelerate the slow rate of the randomized power method (without requiring knowledge of the eigengap) into a robust algorithm achieving the optimal rate of convergence.

📄 PDF Abstract BibTeX arXiv:1802.09128

Code (0)

등록된 구현이 없습니다.

Tasks

Riemannian optimization

Similar Papers 제목 키워드 기반

Zeroth-order Riemannian Averaging Stochastic Approximation Algorithms

2023-09-25 · Jiaxiang Li, Krishnakumar Balasubramanian, Shiqian Ma

We present Zeroth-order Riemannian Averaging Stochastic Approximation (\texttt{Zo-RASA}) algorithms for stochastic optimization on Riemannian manifolds. We show that \texttt{Zo-RASA} achieves optimal sample complexities …

Stochastic Optimization

Mirror Descent on Riemannian Manifolds

2026-03-18 · Jiaxin Jiang, Lei Shi, Jiyuan Tan arxiv

Mirror Descent (MD) is a scalable first-order method widely used in large-scale optimization, with applications in image processing, policy optimization, and neural network training. This paper generalizes MD to optimiza…

Stochastic gradient descent on Riemannian manifolds

2011-11-22 · Silvere Bonnabel

Stochastic gradient descent is a simple approach to find the local minima of a cost function whose evaluations are corrupted by noise. In this paper, we develop a procedure extending stochastic gradient descent algorithm…

Riemannian optimization

Riemannian stochastic optimization methods avoid strict saddle points

2023-11-04 · NeurIPS 2023 11

Many modern machine learning applications - from online principal component analysis to covariance matrix identification and dictionary learning - can be formulated as minimization problems on Riemannian manifolds, and a…

Dictionary LearningPolicy Gradient MethodsRiemannian optimizationStochastic Optimization

Riemannian Adaptive Optimization Methods

2018-10-01 · ICLR 2019 5 · Gary Bécigneul, Octavian-Eugen Ganea

Several first order stochastic optimization methods commonly used in the Euclidean domain such as stochastic gradient descent (SGD), accelerated gradient descent or variance reduced methods have already been adapted to c…

Riemannian optimizationStochastic Optimization