AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

DyS: A Framework for Mixture Models in Quantification

André Maletzke, Denis dos Reis, Everton Cherman, Gustavo Batista

AAAI 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Group-robust classification · 44 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
use fewer histogram bins (up to about 20) instead of 10-110 bins (ablation)DyS with histogram distances (Topsøe, Hellinger) · 24 binary datasetsMAE: HD: 95% of best results within 2 to 14 binsImproved
increase test sample size (ablation)DyS · 24 binary datasetsMAEImproved
replace Hellinger distance with Topsøe (tuned bins) (ablation)DyS / HDy · 24 binary datasetsquantification rank / MAEImproved
use parameter-free SORD instead of Hellinger (HDy) (ablation)DyS · 24 binary datasetsquantification errorImproved
use Cosine, Harmonic Mean, Kumar-Hassebrook or Inner Product distances (ablation)DyS · 24 binary datasetsMAE: errors greater than 70% on averageHurt
use ORD distance (insensitive to number of bins) (ablation)DyS · 24 binary datasetsquantification errorMixed

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