AI Paper Atlas

Research map / Label-efficient and robust learning / Noisy labels

Overlap-Robust Decision Boundary Learning for Within-Network Classification

Sharad Nandanwar, M. N. Murty

AAAI 2018 · Label-efficient and robust learning › Noisy labels · approach: Contrastive learning on noisy data · 2 citations (OpenAlex)

Read the paper

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
add homophily regularizer to ridge in structural-loss ReSLMin (ablation)ReSLMin · PubMedMicro F1 (%): 80.23±0.10 -> 80.83±0.04Improved
add homophily regularizer to ridge (ablation)ReSLMin · CoRAMicro F1 (%): 76.43±0.07 -> 77.49±0.07Improved
add homophily regularizer to ridge (ablation)ReSLMin · WikipediaMicro F1 (%): 71.67±0.12 -> 73.65±0.11Improved
add homophily regularizer to ridge (ablation)ReSLMin · AmazonMicro F1 (%): 66.77±0.06 -> 66.75±0.03No clear effect
use homophily regularizer alone instead of combination (ablation)ReSLMin · AmazonMicro F1 (%): 66.75±0.03 -> 67.03±0.04Improved
replace Laplacian regularizer by Ridge+Homophily (ablation)ReSLMin · WikipediaMicro F1 (%): 69.19±0.08 -> 73.65±0.11Improved
replace Laplacian regularizer by Ridge+Homophily (ablation)ReSLMin · AmazonMicro F1 (%): 66.56±0.04 -> 66.75±0.03No clear effect
ReSLMin Ridge+Homophily vs SNBCReSLMin · CoRAMicro F1 (%): 67.39±2.24 -> 77.49±0.07Improved
ReSLMin Ridge+Homophily vs DeepWalkReSLMin · AmazonMicro F1 (%): 31.98±0.03 -> 66.75±0.03Improved
ReSLMin Ridge+Homophily vs SDECReSLMin · PubMedMicro F1 (%): 56.89±0.01 -> 80.83±0.04Improved
ReSLMin Ridge+Homophily vs SCRNReSLMin · WikipediaMicro F1 (%): 71.98±1.22 -> 73.65±0.11Mixed
increase LFR mixing coefficient beyond 0.25 (class overlap), ReSLMin vs SNBCReSLMin · LFR synthetic 10000 nodesaccuracyImproved

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