Neuromorphic engineering takes inspiration from biology to design brain-like systems that are extremely low-power, fault-tolerant, and capable of adaptation to complex and noisy environments. Biological circuits are constructed using low-precision, unreliable and massively parallel neural elements with highly reconfigurable and plastic connections. Recent research in spiking neural networks (SNNs) has demonstrated interesting principles about learning and neural computation that show promise for new models of computation. Understanding and applying these principles to practical problems is only possible if low-cost large-scale spiking neural simulators can be deployed - which is now feasible using the massive parallelism inherent in GPU architectures. I will first review modeling abstractions for neural circuits, and survey the state-of-the art in hardware and software platforms for neuromorphic engineering. Next I describe CARLsim, a SNN simulation environment that exploits the parallel processing power of GPUs to simulate large-scale SNNs and also show some recent modeling experiments performed using the simulator.
Finally, I present an automated parameter tuning framework that utilizes the simulation environment and evolutionary algorithms to tune SNNs. The simulation environment and associated parameter tuning framework presented here can accelerate and ease the development of neuromorphic software and hardware applications, a first step towards the grand challenge of "reverse engineering the brain".
Nikil Dutt is a Chancellor's Professor of CS, EECS, and Cognitive Sciences at the University of California, Irvine. He received his PhD from the University of Illinois at Urbana-Champaign in 1989. His research interests are in embedded and distributed systems, electronic design automation, computer architecture, optimizing compilers, and brain-inspired architectures and computing.
He has received numerous best paper awards and is coauthor of 7 books.
He has served as EiC of ACM TODAES and currently serves as AE of IEEE TVLSI, as well as on the steering, organizing, and program committees of several premier CAD and Embedded System Design. Professor Dutt is a Fellow of the IEEE, an ACM Distinguished Scientist, and recipient of the IFIP Silver Core Award.