Energy-Adaptive Checkpoint-Free Intermittent Inference for Low Power Energy Harvesting Systems
By: Sahidul Islam , Wei Wei , Jishnu Banarjee and more
Potential Business Impact:
Lets smart devices work even with little power.
Deep neural network (DNN) inference in energy harvesting (EH) devices poses significant challenges due to resource constraints and frequent power interruptions. These power losses not only increase end-to-end latency, but also compromise inference consistency and accuracy, as existing checkpointing and restore mechanisms are prone to errors. Consequently, the quality of service (QoS) for DNN inference on EH devices is severely impacted. In this paper, we propose an energy-adaptive DNN inference mechanism capable of dynamically transitioning the model into a low-power mode by reducing computational complexity when harvested energy is limited. This approach ensures that end-to-end latency requirements are met. Additionally, to address the limitations of error-prone checkpoint-and-restore mechanisms, we introduce a checkpoint-free intermittent inference framework that ensures consistent, progress-preserving DNN inference during power failures in energy-harvesting systems.
Similar Papers
Energy-Aware Deep Learning on Resource-Constrained Hardware
Machine Learning (CS)
Saves phone battery for smart computer programs.
E4: Energy-Efficient DNN Inference for Edge Video Analytics Via Early-Exit and DVFS
CV and Pattern Recognition
Saves phone battery by making smart video analysis faster.
Energy Management and Wake-up for IoT Networks Powered by Energy Harvesting
Systems and Control
Saves power for smart devices using their surroundings.