Skip to main content

HOAM THE HO-AM FOUNDATION

menu open
Recognition for the Ennoblement of the Human Sprit in Academics, The Arts, and Human Welfare

2008 Laureates

Hyunjune Sebastian Seung
  • Engineering
  • Hyunjune Sebastian Seung
    Professor, Massachusetts Institute of Technology
  • Education & Work Experience
    1966 Born in New York, NY, USA 1986 B.A., Physics, Harvard University, MA, USA 1990 Ph.D., Physics, Harvard University, MA, USA 1990 ~ 1992 Postdoctoral Fellow, Racah Institute of Physics and Center for Neural Computation, Hebrew University, Jerusalem, Israel 1992 ~ 1998 Member of Technical Staff, AT&T Bell Laboratories, NJ, USA 1998 ~ 2004 Assistant and Associate Professor of Computational Neuroscience and Physics, MIT, MA, USA 2000 ~ Investigator, Howard Hughes Medical Institute, MD, USA 2004 ~ Professor of Computational Neuroscience and Physics, MIT, MA, USA

Achievements

Dr. Hyunjune Sebastian Seung, a world leader in computational neuroscience, has developed mathematical models for engineering applications by investigating the operation of the brain. He has applied mathematical and physical theories to the understanding of neural phenomena, and has led in the development of intelligent computers based on the brain`s mechanisms of information processing. One of his major achievements is the application of Non-negative Matrix Factorization (NMF) to the learning of features, basic building blocks for the acquisition of knowledge. Using NMF, a computer can learn to represent information as linear combinations of basic components. Both the basic components and the weights of their combination are constrained to have values greater than or equal to zero. These constraints imitate the nonnegativity of neural signals in the brain. Much as Lego blocks are assembled into complex shapes, the basic components are used to represent information in intelligent systems for image recognition, speech recognition, and knowledge acquisition. Together with Principal Component Analysis (PCA) and Independent Component Analysis (ICA), NMF is regarded worldwide as one of the main methods for learning features in diverse applications.

Dr. Seung has also devised mathematical theories of biological neural networks based on the engineering principles of control and optimization. He has advanced models of the neural networks in the brainstem that control eye movements and a ″hedonistic″ theory of neurons and synapses, that is, how they learn to optimize reward just as ″selfish″ genes optimize reproductive fitness in evolutionary theory. Recently Dr. Seung has focused on inventing new technologies for the emerging field of `connectomics`. He is developing computer algorithms that can reconstruct the shapes of neurons and identify synapses in 3-dimensional nanoscale images of the brain. Such technologies are aimed at mapping the brain`s neural networks, providing information about brain structure that will aid in understanding brain function, development, and disease.

Previous Page