Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Convolutional Gaussian 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 |
|---|---|---|---|
| Learn location-specific weights for convolutional patch responses (ablation) | Translation-invariant convolutional GP · MNIST | Error (%): 2.08% -> 1.22% | Helped |
| Learn location-specific weights for convolutional patch responses (ablation) | Translation-invariant convolutional GP · MNIST | NLPP: 0.077 -> 0.048 | Helped |
| Add an RBF component (ablation) | Weighted convolutional GP · MNIST | Error (%): 1.22% -> 1.17% | Helped |
| Add an RBF component (ablation) | Weighted convolutional GP · MNIST | NLPP: 0.048 to 0.039 | Helped |
| Use convolutional structure instead of an RBF kernel | Gaussian process image classifier · CIFAR-10 | Test error: 48.6% -> 35.4% | Helped |
| Use a translation-invariant convolutional kernel instead of RBF | Gaussian process image classifier · MNIST | Error (%): 1.90% -> 2.08% | Hurt |
| Use a translation-invariant convolutional kernel instead of RBF | Gaussian process image classifier · Rectangles | Error: 5.0% -> 1.4% | Helped |
| Use a translation-invariant convolutional kernel instead of RBF | Gaussian process image classifier · Rectangles | NLPP: 0.258 -> 0.055 | Helped |
Closest papers by meaning
- Approximation and Learning with Deep Convolutional Models: a Kernel Perspective (ICLR 2022 · similarity 0.89)
- Unsupervised Visual Domain Adaptation: A Deep Max-Margin Gaussian Process Approach (CVPR 2019 · similarity 0.88)
- Beyond the Mean-Field: Structured Deep Gaussian Processes Improve the Predictive Uncertainties (NeurIPS 2020 · similarity 0.88)
- Graph Convolutional Gaussian Processes (ICML 2019 · similarity 0.88)
- Deep Neural Networks as Point Estimates for Deep Gaussian Processes (NeurIPS 2021 · similarity 0.88)
- Convolutional Deep Kernel Machines (ICLR 2024 · similarity 0.87)
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.