AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

Rank Aggregation from Pairwise Comparisons in the Presence of Adversarial Corruptions

Arpit Agarwal, Shivani Agarwal, Sanjeev Khanna, Prathamesh Patil

ICML 2020 · Adversarial robustness and security › Attacks on text and graphs · approach: Attacks and defenses in sequential settings · 5 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
add LP-based pruning of inconsistent-cycle edges before Accelerated Spectral RankingASR on single-cut-corrupted BTL data · Synthetic, exact probabilities (n=50)L1 error of recovered weightsImproved
add LP-based pruning before ASRASR on single-cut-corrupted BTL data · Synthetic, sampled comparisons (L = log n / eps^2, eps = 5%)L1 error of recovered weightsImproved
increase corruption rate gamma from 5% to 25% for non-robust ASR and HMMASR and HMM · Synthetic dataL1 error of recovered weightsHurt
increase corruption rate gamma towards the upper end of 5%-25% for LP pruningLP Pruning + ASR · Synthetic dataL1 error of recovered weightsMixed
add LP-based pruning before ASRASR · Irish dataset (n=12, p=0.3 subsampled, alpha=0.02)L1 error of recovered weightsImproved
add LP-based pruning before ASRASR · Sushi dataset (n=16, p=0.3 subsampled, alpha=0.02)L1 error of recovered weightsImproved
use robust LP pruning at corruption rates beyond the theoretical limit gamma <= log(np)/(125 log n)LP Pruning + ASR · Real and synthetic dataL1 error of recovered weightsImproved
corrupt only the edges crossing a single cut (scale S up with alpha=0.02)non-robust BTL estimators ASR and HMM · Single cut corruption (Example 1)identifiability of true weightsHurt
restrict per-vertex corruption rate to at most 1/4 - eps on Erdos-Renyi graphsBTL weight identifiability · Erdos-Renyi comparison graphsunique identifiability: 1/4 - epsImproved
increase corruption rate per vertex above 1/4 + epsBTL weight identifiability · Erdos-Renyi comparison graphsunique identifiability: 1/4 + epsHurt
use datasets GIF and Youtube instead of full-ranking datasetsReal-data corruption experiments · GIF, Youtubestructural robustness to corruptionHurt

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