Advanced Intelligent System for Recognition in Real-world, Contents Generation and Knowledge Discovery

Overview

The Harada, Osamu, Kurose and Mukuda Laboratory aims to construct highly intelligent systems capable of understanding the real world, generating content and discovering knowledge by extracting useful information from the real world and combining it with vast amounts of data in cyberspace and powerful computing capabilities.
The laboratory has about 60 members, including both staff and students, and their research utilises all aspects of computer science, including mathematical foundations and robotics.

The main research topics and keywords are as follows.

◆ Mathematical Foundations:
Information theory, machine learning, deep learning, data mining, pattern recognition, probability and statistical theory, time series analysis, causal analysis, learning theory, feature extraction theory

◆Recognition, understanding, thinking:
Big data, computer vision, image recognition and retrieval, 3D information processing, action recognition, multimodal recognition, emotion understanding, natural language processing, speech and music information processing, medical information processing

◆ Content generation:
Natural language description and summarisation of images and videos, image generation from natural language, dialogue systems for chatting with people, discovery of interesting events in the real world and article generation