Frontiers of Information Technology & Electronic Engineering
>> 2019,
Volume 20,
Issue 10
doi:
10.1631/FITEE.1900341
Frontiers of Information Technology & Electronic Engineering
Block coordinate descent with time perturbation for nonconvex nonsmooth problems in real-world studies
Affiliation(s): Department of Oncology, The Second Xiangya Hospital of Central South University, Changsha 410011, China; First Clinical Medical College, University of South China, Hengyang 421001, China; College of Computer, National University of Defense Technology, Changsha 410073, China; less
Accepted: 2019-11-11
Available online: 2019-11-11
Next
Previous
Abstract
The era of big data in healthcare is here, and this era will significantly improve medicine and especially oncology. However, traditional machine learning algorithms need to be promoted to solve such large-scale real-world problems due to a large amount of data that needs to be analyzed and the difficulty in solving problems with nonconvex nonlinear settings. We aim to minimize the composite of a smooth nonlinear function and a block-separable nonconvex function on a large number of block variables with inequality constraints. We propose a novel parallel first-order optimization method, called asynchronous block coordinate descent with (ATP), which adopts a technique that escapes from saddle points and sub-optimal local points. The details of the proposed method are presented with analyses of convergence and iteration complexity properties. Experiments conducted on real-world machine learning problems validate the efficacy of our proposed method. The experimental results demonstrate that enables ATP to escape from saddle points and sub-optimal points, providing a promising way to handle nonconvex optimization problems with inequality constraints employing asynchronous block coordinate descent. The asynchronous parallel implementation on shared memory multi-core platforms indicates that the proposed algorithm, ATP, has strong scalability.