FQTree：提升决策树的细粒度量化与硬件生成
FQTree: Fine-grained Quantization and Hardware Generation of Boosted Decision Trees
FQTree把提升决策树的量化做到硬件级，在FPGA上能省26-57%查找表，精度还不掉，做边缘部署的可以看看。
FQTree是一种针对提升决策树（BDT）的细粒度量化感知训练算法，配套QXGB框架可自动生成低延迟硬件实现。
它采用硬件导向的叶值量化方案，结合全局量化步长与树级移位，实现紧凑的非负整数叶表示、受控裁剪和偏差折叠。该量化在训练过程中逐步应用，使后续树能适应已量化集成模型的误差。
在JSC、MNIST和NID数据集上，相比现有FPGA BDT设计，FQTree将LUT使用量减少26%至57%，同时保持或提升准确率。
FQTree: Fine-grained Quantization and Hardware Generation of Boosted Decision Trees
Boosted decision trees (BDTs) are widely used in latency-critical applications, but efficient hardware deployment remains challenging. Existing designs often rely on uniform or manually tuned fixed-point formats, which can introduce unnecessary hardware cost or accuracy loss. This work presents the FQTree algorithm{https://github.com/ecs-bristol/FQTree} for fine-grained quantization-aware training of BDTs, together with the QXGB framework for automatic hardware generation. FQTree introduces a hardware-oriented leaf-value quantization scheme that uses a global quantization step together with a tree-wise shift, enabling compact non-negative integer leaf representations, controlled clipping/pruning, and bias folding to reduce datapath cost. This work further applies this quantization during boosting so that later trees adapt to the errors of the already-quantized ensemble, and then lowers the trained model into low-latency hardware implementations through a compiler-based flow. Results on JSC, MNIST, and NID show that our method reduces LUT usage by 26-57\% compared with the state-of-the-art FPGA-based BDT designs while matching or improving accuracy.
This report is categorized under AITOP's "Models" desk. Key entities and technical topics under continuous tracking include FQTree, QXGB, 决策树, 量化.
Source attribution: curated from arXiv cs.LG (@Zhiqiang Que, Chang Sun, Haiyang Wang, Dinesh Pamunuwa, Roshan Weerasekera, Qijia Tang, Bakhtiar Zadeh, Wayne Luk, Maria Spiropulu).
