Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Neural Diffusion Processes
Vincent Dutordoir, Alan Saul, Zoubin Ghahramani, Fergus Simpson
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 NDP (diffusion over function marginals with bi-dimensional attention) instead of NP baselines | NDP vs NPs on synthetic GP regression · Squared Exponential and Matern-5/2, D=1-3 | test log-likelihood | Mixed |
| NDP with bi-dimensional attention vs GNP on higher-dimensional GP regression | NDP · Squared Exponential D=2,3 | test log-likelihood: -0.67 vs -0.98 (D=2); -1.16 vs -1.36 (D=3) | Improved |
| Use NDP instead of GNP on Matern-5/2 D=1 regression | NDP · Matern-5/2 D=1 | test log-likelihood: -0.00 vs 0.14 | Hurt |
| Hyperparameter marginalisation via NDP trained on Matern lengthscale prior | NDP · GP samples with lengthscale prior | qualitative lengthscale posterior | Improved |
| Bayesian optimisation with NDP surrogate (no retraining) instead of retrained GPR | NDP · Black-box optimisation, dimensions 3-6 | regret | No clear effect |
Closest papers by meaning
- Neural Flow Diffusion Models: Learnable Forward Process for Improved Diffusion Modelling (NeurIPS 2024 · similarity 0.89)
- Neural Diffusion Models (ICML 2024 · similarity 0.89)
- Scalable Discrete Diffusion Samplers: Combinatorial Optimization and Statistical Physics (ICLR 2025 · similarity 0.87)
- Towards Non-Asymptotic Convergence for Diffusion-Based Generative Models (ICLR 2024 · similarity 0.87)
- Shallow diffusion networks provably learn hidden low-dimensional structure (ICLR 2025 · similarity 0.87)
- Learning diffusion at lightspeed (NeurIPS 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.
