Spintronics based Stochastic Computing for Efficient Bayesian Inference System
This work addresses efficiency limitations in Bayesian inference systems for statistical learning applications, though it appears incremental as it builds on existing spintronics and stochastic computing methods.
The paper tackles the physical bottlenecks of conventional computing platforms for Bayesian inference by proposing a spintronics-based stochastic computing system, achieving significant improvements in power consumption and inference speed as indicated by simulation results.
Bayesian inference is an effective approach for solving statistical learning problems especially with uncertainty and incompleteness. However, inference efficiencies are physically limited by the bottlenecks of conventional computing platforms. In this paper, an emerging Bayesian inference system is proposed by exploiting spintronics based stochastic computing. A stochastic bitstream generator is realized as the kernel components by leveraging the inherent randomness of spintronics devices. The proposed system is evaluated by typical applications of data fusion and Bayesian belief networks. Simulation results indicate that the proposed approach could achieve significant improvement on inference efficiencies in terms of power consumption and inference speed.