Research map / Probabilistic and scientific modeling / Variational and generative sampling
Gradient Boosted Normalizing Flows
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Add 4 boosted components to RealNVP | RealNVP · POWER test | log-likelihood: 0.17 ± .01 -> 0.27 ± 0.01 | Helped |
| Add 4 boosted components to RealNVP | RealNVP · GAS test | log-likelihood: 8.33 ± .14 -> 9.58 ± .04 | Helped |
| Add 4 boosted components to RealNVP | RealNVP · HEPMASS test | log-likelihood: -18.71 ± .02 -> -18.60 ± 0.06 | Helped |
| Add 4 boosted components to RealNVP | RealNVP · MINIBOONE test | log-likelihood: -13.55 ± .49 -> -10.69 ± 0.07 | Helped |
| Add 4 boosted components to RealNVP | RealNVP · BSDS300 test | log-likelihood: 153.28 ± 1.78 -> 154.23 ± 2.21 | Helped |
| Add 4 boosted components to Glow | Glow · POWER test | log-likelihood: 0.17 ± .01 -> 0.24 ± 0.01 | Helped |
| Add 4 boosted components to Glow | Glow · GAS test | log-likelihood: 8.15 ± .40 -> 9.95 ± 0.11 | Helped |
| Add 4 boosted components to Glow | Glow · HEPMASS test | log-likelihood: -18.92 ± .08 -> -17.81 ± 0.12 | Helped |
| Add 4 boosted components to Glow | Glow · MINIBOONE test | log-likelihood: -11.35 ± .07 -> -10.76 ± 0.02 | Helped |
| Add 4 boosted components to Glow | Glow · BSDS300 test | log-likelihood: 155.07 ± .03 -> 154.68 ± 0.34 | Hurt |
| Add 4 boosted RealNVP posterior components | RealNVP VAE · Freyfaces | NLL: 4.62 ± 0.16 -> 4.41 ± 0.01 | Helped |
| Add 4 boosted RealNVP posterior components | RealNVP VAE · MNIST | NLL: 83.36 ± 0.09 -> 82.59 ± 0.03 | Helped |
Closest papers by meaning
- Normalizing Flows are Capable Generative Models (ICML 2025 · similarity 0.90)
- Learning Continuous Normalizing Flows For Faster Convergence To Target Distribution via Ascent Regularizations (ICLR 2023 · similarity 0.89)
- Denoising Normalizing Flow (NeurIPS 2021 · similarity 0.88)
- Kernelised Normalising Flows (ICLR 2024 · similarity 0.87)
- Fast and unified path gradient estimators for normalizing flows (ICLR 2024 · similarity 0.86)
- CDFlow: Building Invertible Layers with Circulant and Diagonal Matrices (NeurIPS 2025 · similarity 0.86)
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.