Early-Exit with Class Exclusion for Efficient Inference of Neural Networks
This addresses the problem of high computational demands for edge devices, but it is incremental as it builds on existing early-exit methods.
The paper tackles the computational inefficiency of deep neural networks on resource-constrained platforms by proposing a class-based early-exit technique that excludes irrelevant classes at intermediate layers, reducing inference cost significantly.
Deep neural networks (DNNs) have been successfully applied in various fields. In DNNs, a large number of multiply-accumulate (MAC) operations are required to be performed, posing critical challenges in applying them in resource-constrained platforms, e.g., edge devices. To address this challenge, in this paper, we propose a class-based early-exit for dynamic inference. Instead of pushing DNNs to make a dynamic decision at intermediate layers, we take advantage of the learned features in these layers to exclude as many irrelevant classes as possible, so that later layers only have to determine the target class among the remaining classes. When only one class remains at a layer, this class is the corresponding classification result. Experimental results demonstrate the computational cost of DNNs in inference can be reduced significantly with the proposed early-exit technique. The codes can be found at https://github.com/HWAI-TUDa/EarlyClassExclusion.