Leon Wang has received the 2005 IEEE Transactions on Neural Networks Outstanding Paper Award (bestowed in 2007) for his paper The Time Dimension for Scene Analysis, published in Vol.16, pp.1401-1426. This award is sponsored annually by the IEEE Computational Intelligence Society and recognizes a single paper published in IEEE Transactions on Neural Networks. Wang will receive this award at the 2008 World Congress on Computational Intelligence, to be held in Hong Kong, June 1-6.
In this paper (posted at http://web.cse.ohio-state.edu/~dwang/papers/Wang.tnn05.pdf), Wang starts with two problems considered by Frank Rosenblatt to be the most challenging to the development of his perceptron theory more than 40 years ago, and points out that the main challenge is the binding problem which refers to how sensory elements in a scene organize into perceived objects. The theme of the paper is that the time dimension is essential for systematically attacking Rosenblatt's challenge. Oscillatory correlation theory is discussed as an adequate representation theory to address the binding problem. Recent advances in understanding oscillatory dynamics have overcome key computational obstacles for the development of oscillatory correlation theory, which in turn have substantially advanced the capability of neural networks for figure-ground separation. In the end, Wang forcefully argues that the time dimension is necessary for versatile computing.