Temporal Decisions: Leveraging Temporal Correlation for Efficient Decisions in Early Exit Neural Networks
This work addresses efficiency and latency issues for embedded and IoT applications, representing an incremental improvement in decision mechanisms for early exit networks.
The paper tackled the challenge of deploying deep learning models on resource-limited embedded devices by proposing new decision mechanisms for Early Exit Neural Networks, achieving up to an 80% reduction in mean operations per inference while maintaining accuracy within 5% of the original model.
Deep Learning is becoming increasingly relevant in Embedded and Internet-of-things applications. However, deploying models on embedded devices poses a challenge due to their resource limitations. This can impact the model's inference accuracy and latency. One potential solution are Early Exit Neural Networks, which adjust model depth dynamically through additional classifiers attached between their hidden layers. However, the real-time termination decision mechanism is critical for the system's efficiency, latency, and sustained accuracy. This paper introduces Difference Detection and Temporal Patience as decision mechanisms for Early Exit Neural Networks. They leverage the temporal correlation present in sensor data streams to efficiently terminate the inference. We evaluate their effectiveness in health monitoring, image classification, and wake-word detection tasks. Our novel contributions were able to reduce the computational footprint compared to established decision mechanisms significantly while maintaining higher accuracy scores. We achieved a reduction of mean operations per inference by up to 80% while maintaining accuracy levels within 5% of the original model. These findings highlight the importance of considering temporal correlation in sensor data to improve the termination decision.