Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Autoregressive Conditional Neural Processes
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 |
|---|---|---|---|
| Deploy ConvCNP autoregressively at test time (ablation) | ConvCNP · Synthetic EQ, d_x,d_y=1 | Norm. KL to truth: 0.41 ± 0.01 -> 0.01 ± 0.00 | Helped |
| Deploy ConvCNP autoregressively at test time (ablation) | ConvCNP · Synthetic EQ, d_x,d_y=2 | Norm. KL to truth: 0.41 ± 0.00 -> 0.03 ± 0.00 | Helped |
| Deploy ConvCNP autoregressively at test time (ablation) | ConvCNP · Synthetic sawtooth, d_x,d_y=1 | Norm. log-lik.: 2.38 ± 0.04 -> 3.60 ± 0.01 | Helped |
| Deploy ConvCNP autoregressively at test time (ablation) | ConvCNP · Synthetic sawtooth, d_x,d_y=2 | Norm. log-lik.: 0.12 ± 0.01 -> 0.38 ± 0.00 | Helped |
| Deploy ConvCNP autoregressively at test time (ablation) | ConvCNP · Synthetic mixture, d_x,d_y=1 | Norm. log-lik.: −0.23 ± 0.04 -> 0.45 ± 0.04 | Helped |
| Deploy ConvCNP autoregressively at test time (ablation) | ConvCNP · Synthetic mixture, d_x,d_y=2 | Norm. log-lik.: −0.85 ± 0.01 -> −0.62 ± 0.01 | Helped |
| Deploy ConvCNP autoregressively at test time (ablation) | ConvCNP · EEG interpolation | Norm. log-likelihood: −1.02 ± 0.01 -> −0.34 ± 0.01 | Helped |
| Deploy ConvCNP autoregressively at test time (ablation) | ConvCNP · EEG reconstruction | Norm. log-likelihood: −2.07 ± 0.03 -> −0.63 ± 0.01 | Helped |
| Deploy ConvCNP autoregressively at test time (ablation) | ConvCNP · Antarctic cloud cover, Gaussian likelihood | Log-lik.: 0.58 ± 0.02 -> 0.88 ± 0.02 | Helped |
| Deploy ConvCNP autoregressively at test time (ablation) | ConvCNP · Antarctic cloud cover, Gaussian likelihood | MAE (%): 13.01 ± 0.16 -> 13.01 ± 0.16 | No clear effect |
| Use beta-categorical rather than Gaussian likelihood with autoregressive ConvCNP (ablation) | ConvCNP (AR) · Antarctic cloud cover | Log-lik.: 0.88 ± 0.02; 1.27 ± 0.02 | Mixed |
| Deploy ConvCNP autoregressively with the multiscale architecture (ablation) | ConvCNP · Temperature downscaling with stations | Norm. log-lik.: −1.55 ± 0.01 -> −1.31 ± 0.01 | Helped |
Closest papers by meaning
- Neural Conditional Probability for Uncertainty Quantification (NeurIPS 2024 · similarity 0.86)
- Rényi Neural Processes (ICML 2025 · similarity 0.86)
- Autoencoding Conditional Neural Processes for Representation Learning (ICML 2024 · similarity 0.86)
- Flow Matching Neural Processes (NeurIPS 2025 · similarity 0.86)
- Hyper Hawkes Processes: Interpretable Models of Marked Temporal Point Processes (NeurIPS 2026 · similarity 0.85)
- Zero-Shot Conditioning of Score-Based Diffusion Models by Neuro-Symbolic Constraints (AAAI 2025 · 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.