Preprint
Aug 2026
FQTree: Fine-grained Quantization and Hardware Generation of Boosted Decision Trees
This work presents the FQTree algorithm for fine-grained quantization-aware training of BDTs, together with the QXGB framework for automatic hardware generation, and applies this quantization during boosting so that later trees adapt to the errors of the already-quantized ensemble.
Zhiqiang Que, Chang Sun, Haiyan Wang et al.
· 0 citations