#ensemble methods
9 papers · 1 filter
Semi-Supervised Learning for Molecular Graphs via Ensemble Consensus
Rasmus Tirsgaard, Laurits Fredsgaard, Marisa Wodrich +2
The paper proposes a semi-supervised learning approach for molecular graph data that uses an ensemble consensus objective to improve prediction accuracy, robustness, and calibratio…
An Interval-Score ROC Curve for Assessment, Calibration and Ensembling of Probabilistic Forecasts
Simone Milanesi, Marco Capelletti, Flavio Bobba +1
The paper proposes the Interval-Score ROC (IS-ROC) curve as a graphical tool for evaluating probabilistic forecasts, and introduces geometry‑based calibration and ensemble methods…
IGME: Efficient Chained Method Ensemble for Transferable Semantic Segmentation Attacks
Mengqi He, Jing Zhang
The paper proposes IGME, an efficient method that chains attack components to generate transferable adversarial perturbations for semantic segmentation using only a single source m…
Dynamical Low-Rank Filters for Data Assimilation
Yoshihito Kazashi, Youssef Marzouk, Fabio Nobile +1
The paper introduces dynamical low-rank (DLR) filters for data assimilation, deriving methods that jointly minimize mean and covariance errors and extending them to Kalman‑Bucy, en…
Do You Really Need to Pretrain Q-Functions for Online RL Fine-Tuning?
Perry Dong, Ron Polonsky, Dorsa Sadigh +2
The paper investigates whether pretraining Q-functions is beneficial when fine‑tuning a pretrained policy in online reinforcement learning, finding that naive Q‑function pretrainin…
Feature Bagging Provides Stability
Yuheng Ma, Qiang Sun
The paper investigates how aggregating models trained on random subsets of features (feature bagging) affects algorithmic stability, introducing a new metric called feature instabi…