Adaptive Acceleration of Inference Services at the Network Edge | IEEE Conference Publication | IEEE Xplore