Parallel coordinate descent for the Adaboost problem

Abstract

We design a randomised parallel version of Adaboost based on previous studies on parallel coordinate descent. The algorithm uses the fact that the logarithm of the exponential loss is a function with coordinate-wise Lipschitz continuous gradient, in order to define the step lengths. We provide the proof of convergence for this randomised Adaboost algorithm and a theoretical parallelisation speedup factor. We finally provide numerical examples on learning problems of various sizes that show that the algorithm is competitive with concurrent approaches, especially for large scale problems.

7 pages, 3 figures, extended version of the paper presented to ICMLA'13

References in corpus (4)

• Mini-Batch Primal and Dual Methods for SVMs
• Parallel Coordinate Descent Methods for Big Data Optimization
• Smooth minimization of nonsmooth functions with parallel coordinate descent methods
• Parallel coordinate descent for the Adaboost problem

Cited by in corpus (7)

• Distributed Coordinate Descent Method for Learning with Big Data
• Stochastic Dual Ascent for Solving Linear Systems
• Smooth minimization of nonsmooth functions with parallel coordinate descent methods
• On Optimal Probabilities in Stochastic Coordinate Descent Methods
• Parallel coordinate descent for the Adaboost problem
• Sketch and Project: Randomized Iterative Methods for Linear Systems and Inverting Matrices
• Separable Approximations and Decomposition Methods for the Augmented Lagrangian