AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

PSEO: Optimizing Post-hoc Stacking Ensemble Through Hyperparameter Tuning

Beicheng Xu, Wei Liu, Keyao Ding, Yupeng Lu, Bin Cui

AAAI 2026 · Learning theory and algorithms › Combinatorial and search optimization · approach: Multi-objective optimization · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
tune ensemble size and diversity weight instead of best fixed pair (ablation)PSEO base model selection · 80 OpenML tasksaverage test rank: 12.3 -> 8.8Improved
add diversity weight (omega > 0) in subset selection (ablation)stacking with BQP selection · 80 OpenML tasksaverage test rank: 17.8-20.4 (omega=0) -> ~12-14Improved
use BQP-based selection instead of ALL (ablation)stacking · 80 OpenML taskswins out of 30 configs: 18 of 30 winsImproved
use BQP-based selection instead of BEST (ablation)stacking · 80 OpenML taskswins out of 30 configs: 29 of 30 winsImproved
increase dropout rate from 0 to 0.4 (ablation)ES blender stacking · 80 OpenML tasksmax weight proportion: decreasing trend (no numbers)Improved
add Retain mechanism to deep stacking (ablation)multi-layer stacking · 80 OpenML tasksfeature quality across layersImproved
use PSEO instead of AutoGluon- strategyPSEO · 80 OpenML tasks (VolcanoML pool)average test rank: 6.19 -> 2.96Improved
use PSEO with AutoGluon base poolPSEO · 80 OpenML tasks (AutoGluon pool)average test rank: 1.90 -> 1.36Improved
stack all models instead of best per algorithm (ablation)stacking baselines · 80 OpenML tasksaverage test rank: ALL-Linear-L1 6.36 vs BEST-Linear-L1 8.69Improved
use stacking instead of ensemble selection (ablation)post-hoc ensemble · 80 OpenML tasksaverage test rank: ES 7.78 -> ALL-ES-L2 6.25Improved
directly search ensemble jointly with models (Autostacker, EO)one-step ensemble learning · 80 OpenML tasksaverage test rank: Single-Best 10.79 vs EO 10.36, Autostacker 9.69Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy