Research map / Probabilistic and scientific modeling / Variational and generative sampling
Flows for simultaneous manifold learning and density estimation
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 |
|---|---|---|---|
| separate manifold and density updates | manifold-learning flow · Mixture model on a polynomial surface | Posterior MMD: 0.131 -> 0.020 | Helped |
| make the manifold independent of the conditioning parameter (ablation) | PIE · Mixture model on a polynomial surface | Mean distance from manifold: 0.035 -> 0.006 | Helped |
| make the manifold independent of the conditioning parameter (ablation) | PIE · Mixture model on a polynomial surface | Posterior MMD: 0.131 -> 0.075 | Helped |
| train the manifold first, then the density (ablation) | M/D manifold-learning flow · Mixture model on a polynomial surface | Posterior MMD: 0.020 -> 0.017 | Helped |
| use a separate encoder (ablation) | M/D manifold-learning flow · Mixture model on a polynomial surface | Posterior MMD: 0.020 -> 0.030 (alternating); 0.017 -> 0.007 (sequential) | Mixed |
| train with simultaneous manifold and likelihood loss (ablation) | manifold-learning flow · Mixture model on a polynomial surface | Posterior MMD: 0.020 -> 0.026 | Hurt |
| train with Sinkhorn divergence (ablation) | manifold-learning flow · Mixture model on a polynomial surface | Posterior MMD: 0.020 -> 0.134 | Hurt |
| alternate adversarial and likelihood training (ablation) | manifold-learning flow · Mixture model on a polynomial surface | Posterior MMD: 0.020 -> 0.051 | Hurt |
| use a lower-dimensional learned manifold rather than an ambient flow | generated samples on the polynomial surface · Mixture model on a polynomial surface | Mean distance from manifold: 0.005 -> 0.002 | Helped |
| use a lower-dimensional learned manifold rather than an ambient flow | inference on θ · Mixture model on a polynomial surface | Posterior MMD: 0.071 -> 0.020 | Helped |
Closest papers by meaning
- Tractable Density Estimation on Learned Manifolds with Conformal Embedding Flows (NeurIPS 2021 · similarity 0.92)
- Learning Distributions on Manifolds with Free-Form Flows (NeurIPS 2024 · similarity 0.90)
- Categorical Flow Matching on Statistical Manifolds (NeurIPS 2024 · similarity 0.90)
- Rectifying Categorical Flows on Statistical Manifolds for One-Step Generation (NeurIPS 2026 · similarity 0.88)
- Lifting Architectural Constraints of Injective Flows (ICLR 2024 · similarity 0.87)
- Generative Modeling on Manifolds Through Mixture of Riemannian Diffusion Processes (ICML 2024 · 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.