Book
Open access
Aug 2026
Efficient Piecewise-Linear Embeddings for Deep Tabular Regression by Guided Breakpoint Allocation
GBDT-Guided Piecewise-Linear (GGPL) embeddings are introduced, which leverage GBDTs to provide a strong data-driven prior for breakpoint placement and fine-tune these breakpoints via gradient descent, which establishes GGPL as an effective numerical embedding for deep tabular regression.
Min-Kook Suh, Moonjung Eo, Kyungeun Lee et al.
· Proceedings of the 32nd ACM... · 0 citations