mlsanityData to 7 Oct 2026

MnasNet: Platform-Aware Neural Architecture Search for Mobile

arXiv 1807.11626 ↗First posted 31 Jul 2018cs.CV cs.LGcited by 937

Not on the map. Why?

Abstract (arXiv)

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

Key facts. Cited by 937 papers in the map (data through 7 Oct 2026). First posted on arXiv 31 Jul 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2018: 1; 2019: 0; 2020: 0; 2021: 299; 2022: 229; 2023: 147; 2024: 115; 2025: 87; 2026: 59.

937 papers.

Sort:

By area:

  1. 2022A ConvNet for the 2020scited by 3,545
  2. 2021EfficientNetV2: Smaller Models and Faster Trainingcited by 1,127
  3. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 798
  4. 2021MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformercited by 636
  5. 2021CoAtNet: Marrying Convolution and Attention for All Data Sizescited by 566
  6. 2023Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectivescited by 477
  7. 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 464
  8. 2021NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Videocited by 241
  9. 2022EfficientFormer: Vision Transformers at MobileNet Speedcited by 240
  10. 2021Coordinate Attention for Efficient Mobile Network Designcited by 239
  11. 2021LocalViT: Analyzing Locality in Vision Transformerscited by 228
  12. 2021Co-Scale Conv-Attentional Image Transformerscited by 224
38 more
  1. 2021Parametric Contrastive Learningcited by 217
  2. 2021Accelerating Sparse Deep Neural Networkscited by 213
  3. 2021AutoFormer: Searching Transformers for Visual Recognitioncited by 186
  4. 2022QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantizationcited by 175
  5. 2023EfficientViT: Memory Efficient Vision Transformer with Cascaded Group Attentioncited by 172
  6. 2021LogME: Practical Assessment of Pre-trained Models for Transfer Learningcited by 168
  7. 2021Neural Architecture Search on ImageNet in Four GPU Hours: A Theoretically Inspired Perspectivecited by 168
  8. 2021Primer: Searching for Efficient Transformers for Language Modelingcited by 162
  9. 2022Compute Trends Across Three Eras of Machine Learningcited by 161
  10. 2023FastViT: A Fast Hybrid Vision Transformer using Structural Reparameterizationcited by 161
  11. 2021Revisiting ResNets: Improved Training and Scaling Strategiescited by 158
  12. 2022On-Device Training Under 256KB Memorycited by 158
  13. 2021Efficient Deep Learning: A Survey on Making Deep Learning Models Smaller, Faster, and Bettercited by 153
  14. 2024EfficientVMamba: Atrous Selective Scan for Light Weight Visual Mambacited by 146
  15. 2023Neural Architecture Search: Insights from 1000 Paperscited by 136
  16. 2021MoViNets: Mobile Video Networks for Efficient Video Recognitioncited by 135
  17. 2023Run, Don't Walk: Chasing Higher FLOPS for Faster Neural Networkscited by 129
  18. 2023Similarity of Neural Network Models: A Survey of Functional and Representational Measurescited by 128
  19. 2021ForgeryNet: A Versatile Benchmark for Comprehensive Forgery Analysiscited by 124
  20. 2023RepViT: Revisiting Mobile CNN From ViT Perspectivecited by 123
  21. 2022Class-Incremental Continual Learning into the eXtended DER-versecited by 121
  22. 2022EdgeNeXt: Efficiently Amalgamated CNN-Transformer Architecture for Mobile Vision Applicationscited by 121
  23. 2022EdgeViTs: Competing Light-weight CNNs on Mobile Devices with Vision Transformerscited by 116
  24. 2021Split Computing and Early Exiting for Deep Learning Applications: Survey and Research Challengescited by 114
  25. 2022MobileOne: An Improved One millisecond Mobile Backbonecited by 109
  26. 2021Accelerated Sparse Neural Training: A Provable and Efficient Method to Find N:M Transposable Maskscited by 103
  27. 2022NeRFusion: Fusing Radiance Fields for Large-Scale Scene Reconstructioncited by 99
  28. 2021The Efficiency Misnomercited by 98
  29. 2024MobileNetV4 -- Universal Models for the Mobile Ecosystemcited by 98
  30. 2021A Survey of Modern Deep Learning based Object Detection Modelscited by 97
  31. 2021ResLT: Residual Learning for Long-tailed Recognitioncited by 96
  32. 2023Rethinking Model Ensemble in Transfer-based Adversarial Attackscited by 88
  33. 2021MCUNetV2: Memory-Efficient Patch-based Inference for Tiny Deep Learningcited by 87
  34. 2023LLMatic: Neural Architecture Search via Large Language Models and Quality Diversity Optimizationcited by 86
  35. 2021P2T: Pyramid Pooling Transformer for Scene Understandingcited by 84
  36. 2021A Study of Face Obfuscation in ImageNetcited by 83
  37. 2021HW-NAS-Bench:Hardware-Aware Neural Architecture Search Benchmarkcited by 83
  38. 2022SimpleRecon: 3D Reconstruction Without 3D Convolutionscited by 83

50 most cited of 937 All 937 (JSON)