Research map / Architectures and efficiency / CNN design
PDO-eS2CNNs: Partial Differential Operator Based Equivariant Spherical CNNs
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 |
|---|---|---|---|
| Use PDO-eS2CNN large instead of HexRUNet-C | Spherical MNIST classification · Spherical MNIST N/N | accuracy (%): 99.45 -> 99.60 | Helped |
| Use PDO-eS2CNN large instead of Icosahedral CNN | Spherical MNIST classification · Spherical MNIST R/R | accuracy (%): 99.31 -> 99.45 | Helped |
| Use PDO-eS2CNN large instead of FFS2CNN | Spherical MNIST classification · Spherical MNIST N/R | accuracy (%): 96 -> 94.25±0.29 | Hurt |
| Use PDO-eS2CNN instead of HexRUNet | Omnidirectional segmentation · 2D-3D-S | mAcc: 58.6 -> 60.4±1.0 | Helped |
| Use PDO-eS2CNN instead of HexRUNet | Omnidirectional segmentation · 2D-3D-S | mIoU: 43.3 -> 44.6±0.4 | Helped |
| Use PDO-eS2CNN instead of MLP/Random CM | QM7 energy prediction · QM7 | RMSE: 5.96±0.48 -> 3.78±0.07 | Helped |
| Use PDO-eS2CNN instead of S2CNN | QM7 energy prediction · QM7 | RMSE: 8.47 -> 3.78±0.07 | Helped |
| Use PDO-eS2CNN instead of FFS2CNN | QM7 energy prediction · QM7 | RMSE: 7.97 -> 3.78±0.07 | Helped |
Closest papers by meaning
- PDO-eConvs: Partial Differential Operator Based Equivariant Convolutions (ICML 2020 · similarity 0.88)
- Bridging Equivariant GNNs and Spherical CNNs for Structured Physical Domains (NeurIPS 2025 · similarity 0.88)
- Spherical CNNs (ICLR 2018 · similarity 0.88)
- Equivariance versus Augmentation for Spherical Images (ICML 2022 · similarity 0.88)
- Scalable and Equivariant Spherical CNNs by Discrete-Continuous (DISCO) Convolutions (ICLR 2023 · similarity 0.88)
- O$n$ Learning Deep O($n$)-Equivariant Hyperspheres (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.