FOX-NAS: Fast, On-device and Explainable Neural Architecture Search
This work addresses the problem of slow neural architecture search for researchers and practitioners needing efficient on-device deployment, though it is incremental as it builds on existing One-Shot approaches.
The paper tackles the inefficiency of One-Shot neural architecture search methods by proposing FOX-NAS, which uses fast and explainable predictors to discover models that match MobileNetV2 and EfficientNet-Lite0 accuracy with 240% and 40% less latency on edge CPUs.
Neural architecture search can discover neural networks with good performance, and One-Shot approaches are prevalent. One-Shot approaches typically require a supernet with weight sharing and predictors that predict the performance of architecture. However, the previous methods take much time to generate performance predictors thus are inefficient. To this end, we propose FOX-NAS that consists of fast and explainable predictors based on simulated annealing and multivariate regression. Our method is quantization-friendly and can be efficiently deployed to the edge. The experiments on different hardware show that FOX-NAS models outperform some other popular neural network architectures. For example, FOX-NAS matches MobileNetV2 and EfficientNet-Lite0 accuracy with 240% and 40% less latency on the edge CPU. FOX-NAS is the 3rd place winner of the 2020 Low-Power Computer Vision Challenge (LPCVC), DSP classification track. See all evaluation results at https://lpcv.ai/competitions/2020. Search code and pre-trained models are released at https://github.com/great8nctu/FOX-NAS.