论文标题
混合精液神经网络：一项调查
Mixed-Precision Neural Networks: A Survey
论文作者
论文摘要
混合精确的深神经网络实现了硬件部署所需的能源效率和吞吐量，尤其是在资源有限的情况下，而无需牺牲准确性。但是，不容易找到保留精度的最佳每层位精度，尤其是在创建巨大搜索空间的大量模型，数据集和量化技术的情况下。为了解决这一困难，最近出现了一系列文献，并且已经提出了一些有希望的准确性结果的框架。在本文中，我们首先总结文献中通常使用的量化技术。然后，我们对混合精液框架进行了详尽的调查，该调查是根据其优化技术进行分类的，例如增强学习和量化技术，例如确定性舍入。此外，讨论了每个框架的优势和缺点，我们在其中呈现并列。我们终于为将来的混合精液框架提供了指南。
Mixed-precision Deep Neural Networks achieve the energy efficiency and throughput needed for hardware deployment, particularly when the resources are limited, without sacrificing accuracy. However, the optimal per-layer bit precision that preserves accuracy is not easily found, especially with the abundance of models, datasets, and quantization techniques that creates an enormous search space. In order to tackle this difficulty, a body of literature has emerged recently, and several frameworks that achieved promising accuracy results have been proposed. In this paper, we start by summarizing the quantization techniques used generally in literature. Then, we present a thorough survey of the mixed-precision frameworks, categorized according to their optimization techniques such as reinforcement learning and quantization techniques like deterministic rounding. Furthermore, the advantages and shortcomings of each framework are discussed, where we present a juxtaposition. We finally give guidelines for future mixed-precision frameworks.
