Skip to main content

KCGI Prof. Mochizuki and Mr. Sun's presentation in AI field won the best domestic award!

On November 11, 2020, Dr. Badru Mochizuki, Assistant Professor at The Kyoto College of Graduate Studies for Informatics (KCGI), and Dr. Yimeng Sun's team presented their work at the ITU AI/ML in 5G Challenge, which aims to solve real-world problems by applying machine learning to communication networks, and were selected as one of the top three teams in Japan and won the Grand Prize.Dr. Mochizuki and Mr. Sun are scheduled to participate in the ITU Final Conference, which will be held online worldwide in December.

Dr. Mochizuki and Ms. Sun chose "Network State Estimation by Video Analysis in Real-time Streaming Service" from the three themes proposed by the conference, and estimated the network state (throughput and loss ratio) by using machine learning in real-time.

This theme is a timely research field that could contribute to solving issues specific to modern society, where the use of teleworking systems using webcams such as Zoom is rapidly increasing around the world due to the global spread of the new coronavirus, resulting in extreme congestion.The machine learning model developed by the team achieved 96.3% recognition accuracy from a given real-time video dataset as a network state estimation using deep learning.

The model constructed by Dr. Mochizuki and Mr. Sun uses a neural network, which

  1. (1) Lightweight model: low computation time (can estimate network state of 500 videos in 1 second)
  2. (2) Portable: Not dependent on any specific platform

The advantages of the system, such as the following, were highly evaluated and led to its winning of the Grand Prize.

Machine learning now influences the nature of communication networks, which are the lifelines of society, and this presentation proposes one way to apply machine learning to the networks of the future.This model has the potential to improve communication speed, communication delay, and the communication system itself, and is expected to be used in the future.

2021年のニュース

2020年のニュース

2019年のニュース

2018年のニュース

2017年のニュース

2016年のニュース

2015年のニュース

2014年のニュース

2013年のニュース

2012年のニュース

2011年のニュース

2010年のニュース

2009年のニュース

2008年のニュース

2007年のニュース