Hugging Face Trending Papers

EMA-FS: Accelerating GBDT Training via Gain-Informed Feature Screening

Read the original on Hugging Face Trending Papers →

Gradient Boosted Decision Trees (GBDT), exemplified by LightGBM, spend a dominant fraction of training time -- typically 65-70% -- constructing per-feature histograms. Existing approaches such as random feature subsampling (feature_fraction) discard features without regard for their predictive utility.

Summary generated by The Flow from the publisher's feed. The full article lives at Hugging Face Trending Papers.