AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Likelihood over Estimation: Robust Quadratic Discriminant Analysis for Heavy-Tailed Distributions with Theory and Evidence

Niranjana Ambadi, eugene pinsky

ICML 2026 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Likelihood-free and graphical model estimation

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
replace Gaussian likelihood with surrogate alpha-stable likelihood (mean + Ledoit-Wolf)QDA · Credit Card FraudPR-AUC: 0.488 -> 0.739Improved
replace Gaussian likelihood with surrogate alpha-stable likelihood (mean + Ledoit-Wolf)QDA · HTRU2recall at 95% precision: 0.707 -> 0.796Improved
replace Gaussian likelihood with stable likelihoodQDA · Ionosphereaccuracy (%): 87.46 -> 91.74Improved
use Tyler's M-estimator for dispersion instead of Ledoit-Wolf shrinkage (ablation)Stable-QDA · HTRU2accuracy (%): 97.83 -> 93.85Hurt
use Tyler's M-estimator for dispersion instead of Ledoit-Wolf shrinkage (ablation)Stable-QDA · Credit Card FraudPR-AUC: 0.739 -> 0.437Hurt
use Tyler's M-estimator on Ionosphere (ablation)Stable-QDA · Ionosphereaccuracy (%): 91.74 -> 92.03No clear effect
use spatial median instead of sample mean with Ledoit-Wolf (ablation)Stable-QDA · HTRU2accuracy (%): 97.83 -> 97.81No clear effect
use a fixed alpha 1.5 instead of the true tail index (ablation)Stable-QDA · Synthetic stable dataaccuracy (%): matches oracle within 0.1%No clear effect
estimate alpha with McCulloch's method instead of fixed 1.5 (ablation)Stable-QDA · Synthetic stable dataaccuracy (%): 76.4 -> 76.4 averageNo clear effect
use per-class alpha instead of shared alpha (ablation)Stable-QDA · Synthetic stable data (alpha 1.0 and 1.8)accuracy (%): 77.1 -> 74.1Hurt
use Gaussian QDA on training data contaminated with 20% outliersGaussian QDA · Stable alpha 1.8 base dataaccuracy (%): 91.0 -> 73.6Hurt
use Tyler's M-estimator under adversarial outliers (ablation)Stable-QDA · Gaussian base, 10% adversarial outliersaccuracy (%): 93.1 -> 90.8Hurt

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