• PI: Assoc. Prof. Than Quang Khoat
  • Time: 9/2019-9/2022
  • Funding: Vingroup Innovation Foundation
  • Code: VINIF.2019.DA18
  • Main areas: Machine Learnning, Big Data, Human Knowledge, Infinite Stream


The core goal of this project is to develop methods that make it possible for computers not only to learn continuously from data but also to make good use of human knowledge in the learning process. Such methods are capable of working with big data or infinite streams of data. This topic is part of a series of research collaborations between Hanoi University of Science and Technology, Kyoto University (Japan) and the University of Oregon (USA).

More details


This project is expected to have a range of impacts:

  1. In theory: the project seeks a new way to solve some of the fundamental challenges in helping computers learn continuously from data streams. The project explores new ways to encode knowledge from humans or from the outside into a model so that machines will better learn and infer from (infinite) data sequences. Finally, the project’s studies provide effective solutions to some of the core problems of future Artificial Intelligence.
  2. In practice: many applications must operate on continuous incoming data streams. Meanwhile, human knowledge is ubiquitous and provides an excellent resource to cope well with many challenges. The studies in this project will merge the two in a unified way to increase the performance of the machine learning model when applied in many areas, such as online advertising, recommendation systems, genetic analysis, standard. Diagnostic…
  3. Education: The project will contribute to training many PhD students, graduate students, and undergraduate students.