Acceleration of Probabilistic Reasoning Through Custom Processor Architecture

Nimish Shah1,a, Laura I. Galindez Olascoaga1,b, Wannes Meert2 and Marian Verhelst1,c
1MICAS, Department of Electrical Engineering, KU Leuven, Belgium
animish.shah@esat.kuleuven.be
blaura.galindez@esat.kuleuven.be
cmarian.verhelst@esat.kuleuven.be
2DTAI, Department of Computer Science, KU Leuven, Belgium
wannes.meert@cs.kuleuven.be

ABSTRACT


Probabilistic reasoning is an essential tool for robust decision-making systems because of its ability to explicitly handle real-world uncertainty, constraints and causal relations. Consequently, researchers are developing hybrid models by combining Deep Learning with probabilistic reasoning for safety-critical applications like self-driving vehicles, autonomous drones, etc. However, probabilistic reasoning kernels do not execute efficiently on CPUs or GPUs. This paper, therefore, proposes a custom programmable processor to accelerate sum-product networks, an important probabilistic reasoning execution kernel. The processor has an optimized datapath architecture and memory hierarchy optimized for sum-product networks execution. Experimental results show that the processor, while requiring fewer computational and memory units, achieves a 12x throughput benefit over the Nvidia Jetson TX2 embedded GPU platform.

Keywords: Sum-product networks, Arithmetic circuits, Custom processor, Probabilistic reasoning, GPU, acceleration



Full Text (PDF)