Research map / Learning theory and algorithms / Prediction and decision losses

Sensitivity Verification for Additive Decision Tree Ensembles

ICLR 2025 · Learning theory and algorithms › Prediction and decision losses · approach: Tree ensemble optimization

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Studies formal verification of sensitivity properties for additive decision tree ensembles such as GBDTs and random forests; the abstract is truncated, so details beyond the stated motivation are unknown.

Problem

Analyzing tree ensemble models (GBDTs, random forests) is harder than single trees; existing exact and approximate techniques exist for questions like robustness, and this work targets sensitivity verification for these ensembles.

Approach

Not stated in the available abstract; the paper extends prior exact/approximate verification techniques for tree ensembles to sensitivity properties of additive ensembles.

Closest papers by meaning

• Data Aware and Scalable Sensitivity Analysis for Decision Tree Ensembles (ICLR 2026 · similarity 0.93)
• Versatile Verification of Tree Ensembles (ICML 2021 · similarity 0.92)
• Robustness Verification of Tree-based Models (NeurIPS 2019 · similarity 0.92)
• Robustness Verification of Multi-Class Tree Ensembles (AAAI 2024 · similarity 0.88)
• Multistage Defer Trees for Hybrid Interpretability: If at First You Can't Succeed, Tree Again (NeurIPS 2026 · similarity 0.88)
• Compressing tree ensembles through Level-wise Optimization and Pruning (ICML 2025 · similarity 0.88)