MnasNet: Platform-Aware Neural Architecture Search for Mobile

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/mnasnet

PDF Abstract CVPR 2019 PDF CVPR 2019 Abstract

Results from the Paper


Ranked #832 on Image Classification on ImageNet (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
Image Classification ImageNet MnasNet-A3 Top 1 Accuracy 76.7% # 832
Number of params 5.2M # 410
Hardware Burden None # 1
Operations per network pass 0.0403G # 1
GFLOPs 0.806 # 94
Image Classification ImageNet MnasNet-A2 Top 1 Accuracy 75.6% # 872
Number of params 4.8M # 394
GFLOPs 0.680 # 80
Image Classification ImageNet MnasNet-A1 Top 1 Accuracy 75.2% # 883
Number of params 3.9M # 376
GFLOPs 0.624 # 73

Methods