On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,657 papers · 148 categories

Trend · papers per month

count % of corpus

2 results for “DRU”

Ranking & Reweighting Improves Group Distributional Robustnesscs.LG

This paper improves model robustness to underrepresented groups using ranking metrics and reweighting.

problem Underrepresented groups suffer from low accuracy in models trained via ERM.

method Proposes Discounted Cumulative Gain (DCG) and Discounted Rank Upweighting (DRU) methods.

result Models trained with DRU show superior generalization to unseen groups.

group robustness ranking metrics reweighting DCG DRU

2023-05-09abs ↗pdf ↗

A Directional Rockafellar-Uryasev Regressionstat.ML

Proposes a neural network model to improve predictions in biased datasets.

problem Reduces bias in predictions from datasets with selection bias.

method Integrates meta information about bias direction and quantity into a neural network model.

result Improves prediction accuracy in electoral polls with biased training data.

bias mitigation meta information neural networks selection bias electoral polls

2024-11-04abs ↗pdf ↗