My research interests broadly cover energy-efficient and reliable computing, from the system level to device level. I am currently working on a project that explores the complementary properties of deep learning and probabilistic inference for making perceptual decisions, where the weaknesses of one can be addressed by the strengths of the other. We are investigating various algorithmic and hardware acceleration approaches that provide effective robot perception in unstructured, natural environments in real time and at efficient energy cost. One goal of the project is to create a general-purpose library of optimized hardware modules for accelerating robot-oriented algorithms.