Can Table Modeling Scale? Rethinking the Tree Model Era Amid Compute Shifts

The article examines how a single NVIDIA H100 GPU delivers roughly 200‑fold more FP16 compute than a 96‑core CPU Hadoop node, explores the "Bitter Lesson" of scaling‑driven AI breakthroughs, and presents large‑scale pretraining experiments that show table and sequence models now exhibit clear scaling laws, challenging the dominance of traditional tree‑based approaches.

FOUNDKMLPStructured Data

0 likes · 10 min read

Can Table Modeling Scale? Rethinking the Tree Model Era Amid Compute Shifts

DataFunTalk

Jan 7, 2023 · Artificial Intelligence

How to Better Leverage Data in Causal Inference

This presentation introduces two recent works from Ant Group that improve causal inference by explicitly using historical control data to reduce selection bias and by fusing heterogeneous multi‑source data, describing the GBCT and WMDL methods, their theoretical foundations, experimental results, and practical applications in finance.

Bias Correctioncausal inferencedata fusion

0 likes · 18 min read

How to Better Leverage Data in Causal Inference

Alimama Tech

May 18, 2022 · Artificial Intelligence

Multiple Boosting Calibration Trees (MBCT): Feature‑Aware Binning for Uncertainty Calibration

The paper introduces Multiple Boosting Calibration Trees, a feature‑aware binning ensemble that uses a new multi‑view calibration error metric and boosting to learn personalized, non‑monotonic calibrations for CTR prediction, achieving lower calibration error and higher click‑through rates and revenue than existing methods in both offline and online tests.

MVCEfeature-aware binningtree models

0 likes · 18 min read

Multiple Boosting Calibration Trees (MBCT): Feature‑Aware Binning for Uncertainty Calibration