Research map / Probabilistic and scientific modeling / Variational and generative sampling
Implicit Generative Copulas
Tim Janke, Mohamed Ghanmi, Florian Steinke
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 |
|---|---|---|---|
| Use IGC instead of vine copula (TLL2) | Copula model · 5D vine copula simulation | mean ISE: 0.2093 -> 0.0697 | Improved |
| Use IGC instead of vine copula (TLL2) | Copula model · 5D vine copula simulation | median ISE: 0.0339 -> 0.045 | Hurt |
| Use IGC instead of Gaussian copula | Copula model · MAGIC | ISE: 0.2415 -> 0.0345 | Improved |
| Use IGC instead of GMMN with sigmoid output | Copula model · Exchange rates | ISE: 0.5791 -> 0.3829 | Improved |
| Use IGC instead of vine copula | Copula model · Exchange rates | ISE: 0.2457 -> 0.3829 | Hurt |
| Use IGC instead of vine copulas | Copula model · MAGIC | ISE: 0.0588 -> 0.0345 | Improved |
| Use IGC instead of GMMN copula | Copula autoencoder · FashionMNIST latent data space | MMD: 0.00341 -> 0.00114 | Improved |
| Use IGC instead of GMMN copula | Copula autoencoder · FashionMNIST latent unit space | MMD: 0.00073 -> 0.00069 | Improved |
| Use IGC instead of GMMN copula | Copula autoencoder · FashionMNIST image space | MMD: 0.00392 -> 0.00426 | No clear effect |
| Use IGC copula on autoencoder latent space instead of VAE | Generative model · FashionMNIST image space | MMD: 0.01316 -> 0.00426 | Improved |
| Model latent dependence with IGC instead of independence copula | Copula autoencoder · FashionMNIST image space | MMD: 0.01912 -> 0.00426 | Improved |
| Use IGC instead of BETA and TLL | Copula density estimator · Bivariate Gaussian mixture copula | ISE | Improved |
Closest papers by meaning
- TACTiS-2: Better, Faster, Simpler Attentional Copulas for Multivariate Time Series (ICLR 2024 · similarity 0.85)
- A Likelihood Based Approach to Distribution Regression Using Conditional Deep Generative Models (ICML 2025 · similarity 0.84)
- HDT: Hierarchical Discrete Transformer for Multivariate Time Series Forecasting (AAAI 2025 · similarity 0.83)
- Discrete Copula Diffusion (ICLR 2025 · similarity 0.83)
- Conditional Diffusion Models are Minimax-Optimal and Manifold-Adaptive for Conditional Distribution Estimation (ICLR 2025 · similarity 0.82)
- Generative Marginalization Models (ICML 2024 · similarity 0.82)
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.
