MnasNet: Platform-Aware Neural Architecture Search for Mobile
Preprint
Abstract
Designing convolutional neural networks (CNN) for mobile devices is challenging because mobile models need to be small and fast, yet still accurate. Although significant efforts have been dedicated to design and improve mobile CNNs on all dimensions, it is very difficult to manually balance these trade-offs when there are so many architectural possibilities to consider. In this paper, we propose an automated mobile neural architecture search (MNAS) approach, which explicitly incorporate model latency into the main objective so that the search can identify a model that achieves a good trade-off between accuracy and latency. Unlike previous work, where latency is considered via another, often inaccurate proxy (e.g., FLOPS), our approach directly measures real-world inference latency by executing the model on mobile phones. To further strike the right balance between flexibility and search space size, we propose a novel factorized hierarchical search space that encourages layer diversity throughout the network. Experimental results show that our approach consistently outperforms state-of-the-art mobile CNN models across multiple vision tasks. On the ImageNet classification task, our MnasNet achieves 75.2% top-1 accuracy with 78ms latency on a Pixel phone, which is 1.8x faster than MobileNetV2 [29] with 0.5% higher accuracy and 2.3x faster than NASNet [36] with 1.2% higher accuracy. Our MnasNet also achieves better mAP quality than MobileNets for COCO object detection. Code is at https://github.com/tensorflow/tpu/tree/master/models/official/mnasnetKeywords
All Related Versions
This publication has 15 references indexed in Scilit:
- ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design2018
- DPP-Net: Device-Aware Progressive Search for Pareto-Optimal Neural Architectures2018
- NetAdapt: Platform-Aware Neural Network Adaptation for Mobile Applications2018
- AMC: AutoML for Model Compression and Acceleration on Mobile Devices2018
- CondenseNet: An Efficient DenseNet Using Learned Group Convolutions2018
- MorphNet: Fast & Simple Resource-Constrained Structure Learning of Deep Networks2018
- Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference2018
- YOLO9000: Better, Faster, Stronger2017
- Deep Residual Learning for Image Recognition2016
- Microsoft COCO: Common Objects in Context2014