Research map / Learning theory and algorithms / Matrix and tensor methods
Adversarial Crowdsourcing Through Robust Rank-One Matrix Completion
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use M-MSR instead of RPCA | rank-one matrix exact recovery · Exact recovery, dimension 10 | running time: 0.46 seconds -> 0.01 seconds | Helped |
| Use M-MSR instead of RPCA | rank-one matrix exact recovery · Exact recovery, dimension 1000 | running time: 80.62 seconds -> 0.42 seconds | Helped |
| Use M-MSR for crowdsourcing with adversarial workers | synthetic crowdsourcing · Varying observation sparsity | prediction error | Helped |
| Increase the number of tasks | M-MSR · Synthetic crowdsourcing | prediction accuracy | Helped |
| Increase the upper bound of normal workers' skill distribution | M-MSR · Synthetic crowdsourcing | prediction error | Helped |
| Vary task-assignment variance | crowdsourcing methods · Synthetic crowdsourcing | prediction error | No clear effect |
| Increase adversary accuracy from 0 to 0.5 | M-MSR · Synthetic crowdsourcing | prediction error: 0 to around 0.35 | Hurt |
| Increase adversary accuracy from 0.5 to 1 | M-MSR · Synthetic crowdsourcing | prediction error: to 0 | Helped |
| Increase interaction-graph sparsity | M-MSR · Synthetic crowdsourcing | prediction error | Helped |
| Project estimated worker skills away from the boundary of the allowed skill range | M-MSR in synthetic experiments · Synthetic crowdsourcing | prediction error | No clear effect |
Closest papers by meaning
- Gradient Descent for Sparse Rank-One Matrix Completion for Crowd-Sourced Aggregation of Sparsely Interacting Workers (ICML 2018 · similarity 0.89)
- Crowdsourcing with Arbitrary Adversaries (ICML 2018 · similarity 0.85)
- Avoiding Imposters and Delinquents: Adversarial Crowdsourcing and Peer Prediction (NeurIPS 2016 · similarity 0.83)
- Symmetric Matrix Completion with ReLU Sampling (ICML 2024 · similarity 0.81)
- Implicit bias of SGD in $L_2$-regularized linear DNNs: One-way jumps from high to low rank (ICLR 2024 · similarity 0.81)
- Robust Mixture Learning when Outliers Overwhelm Small Groups (NeurIPS 2024 · similarity 0.81)
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.