Research map / Probabilistic and scientific modeling / Variational and generative sampling
Calculating Optimistic Likelihoods Using (Geodesically) Convex Optimization
Viet Anh Nguyen, Soroosh Shafieezadeh-Abadeh, Man-Chung Yue, Daniel Huhn, Wolfram Wiesemann
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Replace nominal QDA likelihoods with FR optimistic likelihoods | QDA classifier (FQDA) · Australian | CCR: 80.03 -> 80.68 | No clear effect |
| Replace nominal QDA likelihoods with KL optimistic likelihoods | QDA classifier (KQDA) · Australian | CCR: 80.03 -> 83.68 | Improved |
| Replace nominal QDA likelihoods with FR optimistic likelihoods | QDA classifier (FQDA) · Banknote authentication | CCR: 98.56 -> 99.07 | Improved |
| Replace nominal QDA likelihoods with KL optimistic likelihoods | QDA classifier (KQDA) · Banknote authentication | CCR: 98.56 -> 99.47 | Improved |
| Replace nominal QDA likelihoods with FR optimistic likelihoods | QDA classifier (FQDA) · Cylinder | CCR: 67.10 -> 70.69 | Improved |
| Replace nominal QDA likelihoods with KL optimistic likelihoods | QDA classifier (KQDA) · Cylinder | CCR: 67.10 -> 70.67 | Improved |
| Replace nominal QDA likelihoods with FR optimistic likelihoods | QDA classifier (FQDA) · German credit | CCR: 71.41 -> 74.50 | Improved |
| Replace nominal QDA likelihoods with KL optimistic likelihoods | QDA classifier (KQDA) · German credit | CCR: 71.41 -> 74.60 | Improved |
| Replace nominal QDA likelihoods with FR optimistic likelihoods | QDA classifier (FQDA) · Housing | CCR: 88.54 -> 88.89 | No clear effect |
| Replace nominal QDA likelihoods with KL optimistic likelihoods | QDA classifier (KQDA) · Housing | CCR: 88.54 -> 92.90 | Improved |
| Replace nominal QDA likelihoods with FR optimistic likelihoods | QDA classifier (FQDA) · Ilpd | CCR: 55.18 -> 57.42 | Improved |
| Replace nominal QDA likelihoods with KL optimistic likelihoods | QDA classifier (KQDA) · Ilpd | CCR: 55.18 -> 57.83 | Improved |
Closest papers by meaning
- Optimistic Distributionally Robust Optimization for Nonparametric Likelihood Approximation (NeurIPS 2019 · similarity 0.91)
- Variance-based Regularization with Convex Objectives (NeurIPS 2017 · similarity 0.85)
- Localization, Convexity, and Star Aggregation (NeurIPS 2021 · similarity 0.85)
- Bayesian Nonparametrics Meets Data-Driven Distributionally Robust Optimization (NeurIPS 2024 · similarity 0.83)
- Bayesian Optimization with Fisher Information Geometry: Gradient Bounds and Trust-Region Methods (NeurIPS 2026 · similarity 0.83)
- A Unified Confidence Sequence for Generalized Linear Models, with Applications to Bandits (NeurIPS 2024 · similarity 0.83)
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.
