Towards a Better Distributed Framework for Learning Big Data
Abstract
This project studies how to perform machine learning tasks more efficiently and effectively in a distributed environment by exploring and solving the following problems: 1) design a general decomposition framework that allows easy scaling of the learning algorithms to a distributed environment without re-designing for parallelism, 2) handle several challenges presented in a distributed setting, such as overcoming the insufficient resources on local sites, balancing the communication and computation loading between server and clients, and learning from data without labels.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 23, 2016
- Source ID
- FA23861514013
Entities
People
- Shou-De Lin
Organizations
- Air Force Office of Scientific Research
- National Taiwan University
- United States Air Force