Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
On Convergence of Polynomial Approximations to the Gaussian Mixture Entropy
Caleb Dahlke, Jason Pacheco
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 Huber et al. Taylor approximation with the proposed convergent Taylor series | GMM entropy approximation, 4th order · Scalar two-component GMM, sigma2^2 in (0,1] | entropy approximation accuracy | Improved |
| add Taylor limit extrapolation to the proposed Taylor series (ablation) | Proposed Taylor series · Scalar and 2-D five-component GMM | entropy approximation accuracy | Improved |
| use Legendre polynomial series instead of Taylor series | Proposed polynomial approximations · Scalar GMM, 4th order | entropy approximation accuracy | Improved |
| shift mean of fifth component of five-component 2-D GMM | Huber et al. approximation, 3rd order · Multivariate GMM, mu5=c[1,1], c in [-3,3] | entropy approximation accuracy | Hurt |
| increase dimension from 1 to 50 | Proposed methods vs Huber et al., 2nd order · Two-component GMMs, Monte Carlo baseline L=1000 | accuracy and computation time | No clear effect |
| increase number of components from 1 to 20 | Huber et al. approximation · Two-dimensional GMMs, 2nd order | standard deviation of error | Hurt |
| use Legendre or Huber instead of Taylor or Taylor limit | Proposed series, varying components · Two-dimensional GMMs, 2nd order | accuracy | Improved |
| increase polynomial order | Proposed series and Huber et al. · Three-dimensional two-component GMM | accuracy | Mixed |
| increase polynomial order | Proposed series · Three-dimensional two-component GMM | computation time | No clear effect |
| replace Jensen lower bound entropy with proposed Taylor, Taylor limit or Legendre entropy in NPV, plus Gauss-Hermite cross-entropy | Nonparametric Variational Inference · Mixture Student T target approximated by GMM | KL divergence | Improved |
| increase number of GMM components | NPV vs modified NPV · Mixture Student T target | KL divergence | Mixed |
| use low order polynomial approximation | Modified NPV · Mixture Student T target | KL divergence | Improved |
Closest papers by meaning
- Toward Global Convergence of Gradient EM for Over-Paramterized Gaussian Mixture Models (NeurIPS 2024 · similarity 0.87)
- Mixtures Closest To A Given Measure: A Semidefinite Programming Approach (ICML 2026 · similarity 0.87)
- Trust-Region Variational Inference with Gaussian Mixture Models (JMLR 2020 · similarity 0.87)
- Sublinear Variational Optimization of Gaussian Mixture Models with Millions to Billions of Parameters (NeurIPS 2026 · similarity 0.86)
- Is $\sqrt{d}$ Separation Necessary for Gradient EM to Learn Gaussian Mixtures in High Dimensions? (NeurIPS 2026 · similarity 0.85)
- Gaussian Mixture Models in Hilbert Spaces via Kernel Methods (NeurIPS 2026 · similarity 0.85)
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.
