%0 Conference Paper
%B IEEE Symposium on VLSI Circuits (VLSI)
%D 2020
%T A 3mm2 Programmable Bayesian Inference Accelerator for Unsupervised Machine Perception using Parallel Gibbs Sampling in 16nm
%A Ko, Glenn
%A Chai, Yuji
%A Marco Donato
%A Paul Whatmough
%A Tambe, Thierry
%A Rob Rutenbar
%A David Brooks
%A Gu-Yeon Wei
%K accelerators
%K bayesian inference
%K deep learning
%X This paper describes a 16nm programmable accelerator for unsupervised probabilistic machine perception tasks that performs Bayesian inference on probabilistic models mapped onto a 2D Markov Random Field, using MCMC. Exploiting two degrees of parallelism, it performs Gibbs sampling inference at up to 1380× faster with 1965× less energy than an Arm Cortex-A53 on the same SoC, and 1.5× faster with 6.3× less energy than an embedded FPGA in the same technology. At 0.8V, it runs at 450MHz, producing 44.6 MSamples/s at 0.88 nJ/sample.
%B IEEE Symposium on VLSI Circuits (VLSI)
%G eng
%U https://doi.org/10.1109/VLSICircuits18222.2020.9162784