Parallel Coordinate Descent Methods for Big Data Optimization

Channel:
United States Simons Institute for the Theory of Computing
Subscribers:
68,700
Published on ● Video Link: https://www.youtube.com/watch?v=IQgnstB0n2E


Duration: 54:43
1,970 views
11

Peter Richtarik, University of Edinburgh
Parallel and Distributed Algorithms for Inference and Optimization
http://simons.berkeley.edu/talks/peter-richtarik-2013-10-23



Other Videos By Simons Institute for the Theory of Computing


2013-10-28Large-Scale Generalized Matching
2013-10-28Consensus-Based Distributed Online Prediction and Optimization
2013-10-28Communication Avoiding Algorithms for Linear Algebra and Beyond
2013-10-28Using Data-Oblivious Algorithms for Private Cloud Storage Access
2013-10-28Parallelism in Linear and Mixed Integer Programming
2013-10-28Peeling Algorithms
2013-10-28Provably Efficient Algorithms for Training Deep Networks
2013-10-28Parallelization by Approximation
2013-10-28Randomized Asynchronous Linear Solvers
2013-10-28An Algebraic View of the New Randomized Kaczmarz Linear Solver
2013-10-28Parallel Coordinate Descent Methods for Big Data Optimization
2013-10-28Distributed Learning, Communication Complexity, and Privacy
2013-10-28On the O(1/k) Convergence of the Asynchronous Distributed Alternating Direction Method of Mult...
2013-10-28From Sparse Matrices to Graphs and Back Again
2013-10-28Implementing Randomized Matrix Algorithms in Parallel and Distributed Environments
2013-10-28Communication-Optimal Parallel Algorithm for Strassen's Matrix Multiplication
2013-10-28Big Data on Smallish Machines
2013-10-28Sampling Algorithms to Count Frequent Patterns in Graphs
2013-10-28MLbase: A Distributed Machine Learning System
2013-10-28Extreme Multiclass
2013-10-28Cost Models for Locality and Parallelism


Tags:
Simons Institute
Big Data
UC Berkeley
computer science
theory of computing
Peter Richtarik