{
  "id": "1905.11669",
  "version": "v1",
  "published": "2019-05-28T08:24:58.000Z",
  "updated": "2019-05-28T08:24:58.000Z",
  "title": "CompactNet: Platform-Aware Automatic Optimization for Convolutional Neural Networks",
  "authors": [
    "Weicheng Li",
    "Rui Wang",
    "Zhongzhi Luan",
    "Di Huang",
    "Zidong Du",
    "Yunji Chen",
    "Depei Qian"
  ],
  "categories": [
    "cs.LG",
    "stat.ML"
  ],
  "abstract": "Convolutional Neural Network (CNN) based Deep Learning (DL) has achieved great progress in many real-life applications. Meanwhile, due to the complex model structures against strict latency and memory restriction, the implementation of CNN models on the resource-limited platforms is becoming more challenging. This work proposes a solution, called CompactNet\\footnote{Project URL: \\url{https://github.com/CompactNet/CompactNet}}, which automatically optimizes a pre-trained CNN model on a specific resource-limited platform given a specific target of inference speedup. Guided by a simulator of the target platform, CompactNet progressively trims a pre-trained network by removing certain redundant filters until the target speedup is reached and generates an optimal platform-specific model while maintaining the accuracy. We evaluate our work on two platforms of a mobile ARM CPU and a machine learning accelerator NPU (Cambricon-1A ISA) on a Huawei Mate10 smartphone. For the state-of-the-art slim CNN model made for the embedded platform, MobileNetV2, CompactNet achieves up to a 1.8x kernel computation speedup with equal or even higher accuracy for image classification tasks on the Cifar-10 dataset.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2019-05-28T08:24:58.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "convolutional neural network",
      "platform-aware automatic optimization",
      "compactnet",
      "state-of-the-art slim cnn model",
      "8x kernel computation speedup"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}