The biological brain implements massively parallel computations using a complex architecture that is different from the conventional Von Neumann computing style. Our brain is a low-power, fault-tolerant and high-performance machine! It consumes only about 20W and brain circuits continue to operate as the organism needs even when the circuit (neuron, neuroglia, etc.) is perturbed or died. Our goal in this project is to research and develop an adaptive and reliable neuro-inspired system and platform with on-chip learning and cognitive capabilities to tackle problems in machine learning and robotics. The project is also expected to provide a promising paradigm for building new generations of fault/error-tolerant computing systems
Currently, we are investigating the following issues: the communication network for medium-scale and massively-parallel neuro-inspired chips for adaptive autonomous systems, reconfigurability and adaptability methods, reliability and fault-tolerance, and learning circuits. In addition, lessons gained from this project will also be investigated to optimize power & performance of the conventional computing systems.