Redundancy Techniques for Straggler Mitigation in Distributed Optimization and Learning
From MaRDI portal
Publication:5381126
zbMath1492.90126arXiv1803.05397MaRDI QIDQ5381126
Wotao Yin, Yifan Sun, Can Karakus, Suhas N. Diggavi
Publication date: 7 June 2019
Full work available at URL: https://arxiv.org/abs/1803.05397
restricted isometry propertycoordinate descentdistributed optimizationproximal gradientstraggler mitigation
Related Items (3)
A divide-and-conquer algorithm for distributed optimization on networks ⋮ Redundancy Techniques for Straggler Mitigation in Distributed Optimization and Learning ⋮ A robust multi-batch L-BFGS method for machine learning
Uses Software
Cites Work
- Unnamed Item
- Queueing with redundant requests: exact analysis
- Faster least squares approximation
- Steiner equiangular tight frames
- The smallest eigenvalue of a large dimensional Wishart matrix
- A limit theorem for the norm of random matrices
- Complex Hadamard matrices and equiangular tight frames
- Global Convergence of Online Limited Memory BFGS
- ARock: An Algorithmic Framework for Asynchronous Parallel Coordinate Updates
- Augmented $\ell_1$ and Nuclear-Norm Models with a Globally Linearly Convergent Algorithm
- Randomized Sketches of Convex Programs With Sharp Guarantees
- Randomized Algorithms for Matrices and Data
- Decoding by Linear Programming
- Near-Optimal Signal Recovery From Random Projections: Universal Encoding Strategies?
- Lower bounds on the maximum cross correlation of signals (Corresp.)
- Speeding Up Distributed Machine Learning Using Codes
- Graph Codes for Distributed Instant Message Collection in an Arbitrary Noisy Broadcast Network
- On Orthogonal Matrices
- “Short-Dot”: Computing Large Linear Transforms Distributedly Using Coded Short Dot Products
- Coded Computation Over Heterogeneous Clusters
- Redundancy Techniques for Straggler Mitigation in Distributed Optimization and Learning
- An Asynchronous Parallel Stochastic Coordinate Descent Algorithm
- Orthogonal Matrices with Zero Diagonal
This page was built for publication: Redundancy Techniques for Straggler Mitigation in Distributed Optimization and Learning
