Juyang (John) Weng received the BS degree in computer science from Fudan University, Shanghai, China, in 1982, and M. Sc. and PhD degrees in computer science from the University of Illinois at Urbana-Champaign, 1985 and 1989, respectively.
He is currently a professor of Computer Science and Engineering at Michigan State University, East Lansing. He is also a faculty member of the Cognitive Science Program and the Neuroscience Program at Michigan State University. His research interests lie in the general area of natural intelligence and artificial intelligence, especially the autonomous development of a variety of mental capabilities by robots and animals, including perception, cognition, behaviors, motivation, and abstract reasoning skills. He has published over 250 research articles on related subjects, including task muddiness, intelligence metrics, mental architectures, vision, audition, touch, attention, recognition, autonomous navigation, natural language understanding, and other emergent behaviors.
In 1992 and 1993, he published with N. Ahuja and T. S. Huang, the Cresceptron, whose journal version appeared in 1997. Cresceptron was inspired by Neocognitron which was for recognition of individual characters in uniform background. Cresceptron appears to be the first visual learning program for general objects in complex natural background by directly taking inputs from sensors and effectors and producing outputs to effectors. It also does segmentation. Since then, he has been concentrating on what is later called autonomous mental development, emphasizing autonomous generation of internal representation, including theories, methods and experimental studies.
He proposed a computational brain-mind model (Weng 2010) called Developmental Network (DN), which appears to be the first meant for the 5-chunk scale — development, architecture, area, space and time. He established (Weng 2011) that a Generative DN (GDN) can learn any given Finite Automata (FA) incrementally, immediately, and error-free. However, unlike the handcrafted FA, GDN is grounded in the physical world using emergent internal representations. This theoretical result bridges the well-known divide between symbolic representations and emergent representations in the sense that “neural networks do not abstract well” (Minsky 1991, also stated by Michael I. Jordan at the David Rumelhart Memorial Plenary Talk IJCNN 2011).
He is a coauthor (with T. S. Huang and N. Ahuja) of the book Motion and Structure from Image Sequences (Springer-Verlag, 1993). He is an Editor-in-Chief of International Journal of Humanoid Robotics and an associate editor of the IEEE Transactions on Autonomous Mental Development. He has chaired and co-chaired a few conferences, including the NSF/DARPA funded Workshop on Development and Learning 2000 (1st ICDL), 2nd International Conference on Development and Learning (ICDL 2002), 7th ICDL (2008), 8th ICDL (2009), and the INNS New Directions in Neural Networks Symposia (NNN 2008). He was the Chairman of the Governing Board of the International Conferences on Development and Learning (ICDLs) (2005-2007), chairman of the Autonomous Mental Development Technical Committee of the IEEE Computational Intelligence Society (2004-2005), an associate editor of IEEE Trans. on Pattern Recognition and Machine Intelligence, and an associate editor of IEEE Trans. on Image Processing. He is a fellow of IEEE.
M. Minsky. “Logical versus analogical or symbolic versus connectionist or neat versus scruffy.” AI Magazine, 12:34-51, 1991.
J. Weng. A 5-chunk developmental brain-mind network model for multiple events in complex backgrounds. In Proc. International Joint Conference on Neural Networks, pages 1–8, Barcelona, Spain, July 18-23 2010.
J. Weng, "Three Theorems: Brain-Like Networks Logically Reason and Optimally Generalize," In Proc. International Joint Conference on Neural Networks, San Jose, CA, pp. +1-8, July 31 - August 5, 2011.
To Weng's Home Page: http://web.cps.msu.edu/~weng/