Report Title: Granular Ball Computing: A New Efficient, Robust and Interpretable Artificial Intelligence Theory

Speaker: Professor Shuyin Xia

Affiliation: Chongqing University of Posts and Telecommunications

Time: 16:00-16:40, Saturday, June 7, 2025

Venue: Meeting Room 706, School of Computer Science and Information Engineering, Hefei University of Technology

Report Abstract:

Existing artificial intelligence methods are mainly based on the finest pixel/single granularity, lacking the naturally efficient, robust and interpretable concept description ability of the human brain's "global first" cognition. To this end, based on the multi-granularity cognitive computing theory, granular ball computing is proposed and developed. Based on a coarse-to-fine generation method to simulate the human brain's "global first" cognitive mechanism, this theory uses granular balls of different granularities to cover data samples, realizing adaptive and efficient multi-granularity concept representation of data; and constructs a new computing mode based on granular balls, achieving a more efficient, robust and interpretable computing mode compared with traditional artificial intelligence methods. At present, granular ball computing has attracted extensive attention from many well-known domestic scholars, and also been followed up by renowned scholars from top international universities such as the University of Michigan, the Indian Institute of Technology, and the University of Alberta. This report introduces the relevant research results and latest progress of the granular ball computing theory, mainly including: granular ball classifier, granular ball clustering, granular ball graph network, granular ball reinforcement learning, granular ball large model, granular ball evolutionary computing, granular ball open continual learning, granular ball federated learning, granular ball rough set, granular ball fuzzy set, granular ball three-way decisions, granular ball superpixel and granular ball NLP, etc.

Speaker Profile:

Shuyin Xia, Professor, National Excellent Young Scientist, Vice Dean of the School of Artificial Intelligence and Vice Dean of the Institute of Advanced Interdisciplinary Research, Chongqing University of Posts and Telecommunications, and Deputy Director of the Key Laboratory of Cyberspace Big Data Intelligent Computing of the Ministry of Education. He presides over a number of national key projects such as the National Key R&D Program and the Original Exploration Program of the National Natural Science Foundation. Together with Professors Guoyin Wang and Xinbo Gao, he proposed and developed the granular ball computing theory. His relevant achievements (first/corresponding author) have been published in important artificial intelligence journals and conferences such as IEEE TPAMI, TKDE, TNNLS, TCYB, ICML, AAAI, IJCAI and ICDE. He has won the CCF Natural Science First Prize, the Wu Wenjun Artificial Intelligence Science and Technology Progress First Prize, the Chongqing Natural Science First Prize and the National Teaching Achievement Second Prize. His research directions include granular ball computing, computer vision, machine learning, deep learning, evolutionary computing and swarm intelligence, etc.