Distributed machine learning on public clouds

• Main Author: Tran, Quang Minh
• Other Authors: Ta Nguyen Binh Duong
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Computer science and engineering
• Online Access: http://hdl.handle.net/10356/70512
• Institution: Nanyang Technological University

Today, machine learning is not something strange anymore. The application of machine learning is nearly everywhere in our daily life. Along with the development of this field, the need of machine learning experiments is also on the rise Moreover, big data is also an emerging topic Nowadays, people hear about Internet of Thing or Industry 4.0 every day. In the past, when the model complexity, number of models to run and dataset size are relatively small, machine learning can be easily done within 1 machine. Moreover, the number of computational devices has been increasing significantly. Particularly, services such as Microsoft Azure, Google Cloud or Amazon Web Service can easily provide developers or researchers with a resourceful infrastructure. Therefore, distributed machine learning comes as a solution to increase performance and utilize abundant resource. However, with various types of machine learning framework and public cloud resource, there is nearly no popular system assisting users to do machine learning in distributed manner. Developing a web system aiming to do so should be a great tool for researchers and developers.