Dr. A. Steven Younger has been Research Professor at CASE/JVIC since 2005. Before that he was Lecturer in the Department of Physics, Astronomy and Materials Science at Missouri State.

He is a Missouri State alumnus, graduating in Physics and Math in 1979. He attended graduate school at the University of Utah in Salt Lake City, receiving a M.S. in Instrumentation Physics in 1983. After working in industry as a Software Engineer and Scientific Programmer for several years, he returned to the University of Utah for further study. He received his Ph.D. in Physics in 1996. His dissertation research was on neural networks, an interest that he continues to research today.

Dr. Younger was Principal Investigator for the project "Embedded Intelligence: Migrating PreAct® Symbolic Constructs into Hardware." The project was performed in partership with JVIC Corporate Affiliate Applied Systems Intelligence of Roswell, GA.

Watch a 10 minute streaming video summary of the project.

His current research interests include Neural Networks, computing with special purpose Optical and Electronic hardware, Signal and Image Processing, Speech and Pattern Recognition.

Recent Teaching:

• Phy 354 : Signals and Systems (alias Digital Signal Processing)
• Phy 123, 124: Introduction to Physics
• Phy 203, 204: Foundations of Physics
• Phy 220: Introduction to Digital Logic
• Phy 386, 486: Senior Research Projects.

Research:

Artificial Neural Networks were inspired by the study of Biological Neural Networks – Central Nervous Systems. By using simple models of these complex biological systems, Neural Network researchers are attempting to develop new computers that are more flexible than standard computers, and that can learn from experience instead of requiring complicated programming.

A good introductory reference on neural networks is "Neural Network Architectures" by Judith Dayhoff.

Optical Neural Networks are attempting to do these computations using beams of light instead of electrical signals. Light has several advantages over electronics. Light is fast, it moves 'at the speed of light,' reducing computation time. Light beams can also cross without disturbing each other. This allows for massivly parallel, three dimensional interconnections between components. Optical media, such as photographic film, can store large amounts of information in a small area. Finally, holographic techniques can be used to store and process information, increasing the flexibility of optical computing.

Publications:

A. Steven Younger and Emmett Redd. “Design of an Optical Fixed-Weight Learning Neural Network,” Proceedings of International Joint Conference on Neural Networks, Montreal, Canada, July 31 - August 4, 2005. IEEE 2005 pp. 610-615 PDF

Emmett Redd and A. Steven Younger. “DLP Driven, Learning, Optical Neural Networks,” Presented at Texas Instruments Development Conference. Feb 8-March 2, 2006 Dallas Texas. Site URL: http://www.tidevcon.com/na, paper URL: http://focus.ti.com/lit/ml/sprp506/sprp506.pdf

A. Steven Younger, Sepp Hochreiter and Peter R. Conwell. “Meta-Learning with Backpropagation," Proceedings of International Joint Conference on Neural Networks, Washington D.C. 2001, IEEE 2001 pp. 2001-2006 PDF

Sepp Hochreiter, A. Steven Younger and Peter R. Conwell. “Learning to Learn Using Gradient Descent,” Proceedings of the International Conference on Artificial Neural Networks, Springer Verlag 2001 PDF

A. Steven Younger, Peter R. Conwell and Neil E Cotter, “Fixed-Weight On-Line Learning,” IEEE Transactions on Neural Networks. Vol. 10 No. 2. March 1999 pp. 272-283 PDF

Arthur Steven Younger. Fixed-Weight Learning Recurrent Neural Networks. Ph.D. Dissertation, University of Utah, 1996.