Brain-Mind Workshop Banner

Home | Programs | Committees | Pictures | BMI | BMM | Past BMW

Lateral Inhibition Revisited and its Inspiration to Extreme Learning Machines, Deep Learning and Developmental Networks


Institute of Computing Technology
Chinese Academy of Sciences
Beijing, China


Lateral inhibition (LI) is a classical biological mechanism found in neural systems. In a lot of literature of learning and computing models, LI is referenced as the roles of saliency or edge detection, Winner-Take-All (WTA) or Population-Take-All (PTA). From the point of view of interconnection of neurons, LI is a neuronal structure of Local-Excitation and Global-Inhibition (LEGI). As the central part of LI, Local-Excitation is related to Self Organization Map (SOM) or clustering. As the surrounding part of LI, Global-Inhibition is related to discriminative analysis or classification. In this talk, I will discuss some inspirations form LI to the current frontier topics, such as Extreme Learning Machines (ELM), Deep Learning (DL) and Developmental Networks (DN) and some recent results will be reported.

Short Bio

Jun Miao received the Ph.D. degree in computer science from the Institute of Computing Technology, Chinese Academy of Sciences, in 2005. He is currently an Associated Professor at the Institute of Computing Technology, Chinese Academy of Sciences, Beijing. His research interests include artificial intelligence, neural networks and biologically inspired computer vision. He has published more than 40 research articles in refereed journals and proceedings on object detection, vision neural networks, visual neural information coding, visual perception and cognition. His two main contributions are the technique of Human Face Gravity-Center Template for face detection and the model of Vision Neural Networks for visual search, respectively.

BMW line

Last updated: December 14, 2013