Lab Mission

Our lab builds real systems that operate at the frontiers of computation and physics. We first design biologically-inspired, hardware-friendly algorithms, then we design and architect hardware that runs these algorithms very efficiently. In the end, we use these chips in specific domains such as robotics, biomedical-implants, and anywhere we might need to embed intelligence.

Research

We are studying techniques that are not energy intensive to improve machine learning algorithms

Our studies develop extremely power efficient processors and circuits that can operate under severe resource constraints such as sensors and processors used in the “Internet of Things” IOT. Our group explores hardware and software techniques to enable adaptive and learning algorithms and to be implemented in low-power architectures and circuits that use machine learning algorithms to operate at the limits of energy efficiency. The use of non-von Neumann architectures and both analog-CMOS and Beyond CMOS techniques enables these designs to operate at the extremes of energy efficiency.