Deep-layered machine learning technologies have recently emerged as promising, biologically-inspired cognitive architectures for driving next-generation artificial intelligence applications. Deep learning architecture comprise of hierarchical structures capable of representing diverse, high-dimensional sensory data in a manner that facilitates capturing salient spatiotemporal dependencies in the observations. Advances in the field of machine learning, particularly those made over the past two decades, offer profound insight into the paradigms governing decision making under uncertainty in the mammal brain. From an implementation perspective, integrated circuits fabrication technology continues to improve to a point where billions of neuron-like computing elements can now be realized on a single chip. I argue that these conceptual and implementation building blocks serve as catalysts for the realization of artificial general intelligent systems in the near future.
Itamar Arel is an Associate Professor of Electrical Engineering and Computer Science and Director of the Machine Intelligence Laboratory at the University of Tennessee. He is a co-founder of the Artificial General Intelligence Roadmap initiative, which aims to play a vital role in defining AGI benchmarking and coherence in research focus. During 2000-2003 he was with TeraCross, Inc., a fabless semiconductor company developing Terabit/sec switch fabric integrated circuits, where he held several key positions including chief scientist. His research focus is on high-performance machine learning architectures and algorithms, with emphasis on deep learning architectures, reinforcement learning and decision making under uncertainty. Dr. Arel is a recipient of the US Department of Energy Early Career Principal Investigator (CAREER) award in 2004, and is a senior member of IEEE. He holds a B.S., M.S and Ph.D. degrees in Electrical and Computer Engineering and an M.B.A. degree, all from Ben-Gurion University in Israel.