Decision trees are considered one of the most powerful tools for data classification. Accelerating the decision tree search is crucial for on-the-edge applications that have limited power and latency budget. In this paper, we propose a Content Addressable Memory (CAM) Compiler for Decision Tree (DT) inference acceleration. We propose a novel "adaptive-precision" scheme that results in a compact implementation and enables an efficient bijective mapping to Ternary Content Addressable Memories while maintaining high inference accuracies. In addition, a Resistive-CAM (ReCAM) functional synthesizer is developed for mapping the decision tree to the ReCAM and performing functional simulations for energy, latency, and accuracy evaluations. We study the decision tree accuracy under hardware non-idealities including device defects, manufacturing variability, and input encoding noise. We test our framework on various DT datasets including \textit{Give Me Some Credit}, \textit{Titanic}, and \textit{COVID-19}. Our results reveal up to {42.4\%} energy savings and up to 17.8x better energy-delay-area product compared to the state-of-art hardware accelerators, and up to 333 million decisions per sec for the pipelined implementation.
翻译:决策树被视为数据分类最有力的工具之一。 加快决策树搜索对于电力和潜伏预算有限的前沿应用至关重要。 在本文中, 我们提议为决策树( DT) 推推加速 建立一个内容可处理存储( CAM) 汇编器 。 我们提出一个新的“ 适应性精度” 方案, 其结果为集约执行, 并能够对Ternary 内容可处理的记忆进行高效的双向映射, 同时保持高推力。 此外, 还开发了一个 Resistive- CAM (ReCAM) 功能合成器(ReCAM) 功能合成器, 用于向 RECAM 绘制决策树图, 并进行能源、 延时和精度的功能模拟。 我们研究了硬件非理想下的决定树的准确性, 包括设备缺陷、 制造变异性和输入编码噪音。 我们测试了我们关于各种DT数据集的框架, 包括\ textit{ 给予我一些信用},\ textitriitit{ {Timanic} 和\ text {COD-19} 功能合成合成合成合成合成合成合成合成合成合成合成合成合成合成合成合成系统。 我们的结果显示, 333- creal- creareareal deal deal deal deal deal deal deal deal deal deal deal deal deal deal deal deal deal deal deal deal deal dealmental deal deal deal deal deal deal deal deal deal deal deald 至17.8 ax ax ax ax ax acal decal deal deal decal deal deal decal deal deal deal deal deal deal deal deal deal deal deal deal deal deal deal deald daldaldaldal deal deal dealdaldaldaldaldal deal decal deal deal dealdald a a a a a a a a a d daldal deal de d d d daldaldaldaldaldaldaldaldaldaldaldaldaldaldald