PhoneBit：基于GPU加速的手机二进制神经网络推理引擎
PhoneBit: Efficient GPU-Accelerated Binary Neural Network Inference Engine for Mobile Phones
摘要：
在过去的几年里，深度神经网络（DNN）在计算机视觉和其他领域取得了巨大成功。然而，由于DNN的高计算复杂性，性能和功率约束使得在移动设备上部署DNN仍然具有挑战性。二进制神经网络（BNN）已被证明是一个有前途的解决方案，以实现这一目标，通过使用逐位操作，以取代大多数算术运算。目前，BNN的现有GPU加速实现仅针对桌面平台。由于架构差异，仅仅将这些实现移植到移动设备会产生次优性能，或者在某些情况下是不可能的。在本文中，我们提出了PhoneBit，一个GPU加速的BNN推理引擎，用于基于Android的移动设备，充分利用移动GPU上BNN的计算能力。PhoneBit提供了一组操作员级别的优化，包括本地友好的数据布局，带矢量化的位打包和层集成，以实现高效的二进制卷积。我们还为PhoneBit提供了详细的实现和并行优化，以最佳地利用移动GPU的内存带宽和计算能力。我们评估PhoneBit与AlexNet，YOLOv2 Tiny和VGG16与他们的二进制版本。我们的实验结果表明，PhoneBit可以实现显着的加速比和能源效率相比，国家的最先进的移动设备的框架。
英文摘要：
Over the last years, a great success of deep neural networks (DNNs) has been witnessed in computer vision and other fields. However, performance and power constraints make it still challenging to deploy DNNs on mobile devices due to their high computational complexity. Binary neural networks (BNNs) have been demonstrated as a promising solution to achieve this goal by using bit-wise operations to replace most arithmetic operations. Currently, existing GPU-accelerated implementations of BNNs are only tailored for desktop platforms. Due to architecture differences, mere porting of such implementations to mobile devices yields suboptimal performance or is impossible in some cases. In this paper, we propose PhoneBit, a GPU-accelerated BNN inference engine for Android-based mobile devices that fully exploits the computing power of BNNs on mobile GPUs. PhoneBit provides a set of operator-level optimizations including locality-friendly data layout, bit packing with vectorization and layers integration for efficient binary convolution. We also provide a detailed implementation and parallelization optimization for PhoneBit to optimally utilize the memory bandwidth and computing power of mobile GPUs. We evaluate PhoneBit with AlexNet, YOLOv2 Tiny and VGG16 with their binary version. Our experiment results show that PhoneBit can achieve significant speedup and energy efficiency compared with state-of-the-art frameworks for mobile devices.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
