Towards a Better Distributed Framework for Learning Big Data
Abstract
This work aimed at solving issues in distributed machine learning. The PI's team proposed three directions to work on. First, they designed solutions to speed up the alternating direction method of multipliers (ADMM) for distributed data. Second, they focused on a client-server learning scenario in which an online, semi-supervised learning approach is designed to reduce the communication load. Finally, the team proposed the parallel least-squares policy iteration (parallel LSPI) to parallelize a reinforcement policy learning.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 14, 2017
- Accession Number
- AD1037815
Entities
People
- Shou-De Lin
Organizations
- National Taiwan University