Feature Noise Tuning for Resource Efficient Bayesian Network Classifiers

Emerging portable applications require always-on sensing technologies to continuously monitor the environment and their user’s needs. Yet, the high power consumption that results from this continuous sensing often hampers these systems’ always-on functionality. In this pa- per we propose a hardware-aware Machine Learning scheme that exploits the devices’ ability to trade-off the quality of its sensors versus its power consumption. We introduce a technique that extends Bayesian Network classifiers with hardware description nodes that encode the probabilistic relation between sensory features and their degraded versions. We show how this allows to tune the hardware device’s power consumption versus inference accuracy trade-off space with fine granularity, resulting in oper- ating points that achieve significant power savings at almost no accuracy loss. This is empirically shown on various Machine Learning benchmarking datasets.

ISBN:978-287587047-6

Jaar van publicatie:2018

Toegankelijkheid:Open