AI Paper Atlas

Research map / Learning theory and algorithms / Sample complexity

Spectral Methods Meet EM: A Provably Optimal Algorithm for Crowdsourcing

Yuchen Zhang, Xi Chen, Dengyong Zhou, Michael I. Jordan

JMLR 2016 · Learning theory and algorithms › Sample complexity · approach: Density and property estimation · 247 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 spectral initialization instead of majority-voting initialization for Dawid-Skene EMDawid-Skene EM after one iteration · Synthetic pi=0.2confusion matrix squared error: 5.912 -> 0.972Improved
use spectral initialization instead of majority-voting initializationDawid-Skene EM after 10 iterations · Syntheticconfusion matrix squared error: 0.962 -> 0.962No clear effect
add one EM iteration after the spectral method (ablation)Opt-D&S · Synthetic pi=0.2confusion matrix squared error: 2.084 -> 0.972Improved
add EM iterations after spectral method for dense data (ablation)Opt-D&S · Synthetic pi=1.0confusion matrix squared error: 0.164 -> 0.143Improved
increase sparsity of the data (lower pi)Opt-D&S · Syntheticprediction error (%): 0.01 -> 0.84 -> 7.64 (pi 1.0, 0.5, 0.2)Hurt
use Opt-D&S instead of majority votingmajority voting · Synthetic pi=0.2prediction error (%): 18.85 -> 7.64Improved
use Opt-D&S instead of KOSKOS · TRECerror rate (%): 51.96 -> 29.80Improved
use Opt-D&S instead of MV-D&SMV-D&S · Birderror rate (%): 11.11 -> 10.09Improved
use Opt-D&S instead of MV-D&SMV-D&S · Dogerror rate (%): 16.66 -> 16.89No clear effect
tune the thresholding parameter Delta of spectral initialization (ablation)Opt-D&S 1st iteration · RTE, Dog, Weblabel prediction error: stable after convergence; sensitive after 1st iterationImproved

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