Research map / Architectures and efficiency / CNN design
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 spherical CNN layers instead of planar CNN layers | MNIST digits projected on the sphere · Spherical MNIST NR / R (train non-rotated, test rotated) | accuracy: 0.11 -> 0.94 | Helped |
| Use spherical CNN layers instead of planar CNN layers | MNIST digits projected on the sphere · Spherical MNIST NR / NR (train and test non-rotated) | accuracy: 0.98 -> 0.96 | No clear effect |
| Use spherical CNN layers and spherical molecular signals | QM7 atomization energy regression · QM7 | RMSE: MLP / sorted CM: 16.06; MLP / random CM: 5.96; Spherical CNN: 8.47 | Mixed |
| Train the spherical CNN on the residual rather than predicting energy directly | QM7 atomization energy regression · QM7 | convergence speed and stability | Helped |
Closest papers by meaning
- Efficient Generalized Spherical CNNs (ICLR 2021 · similarity 0.92)
- Scattering Networks on the Sphere for Scalable and Rotationally Equivariant Spherical CNNs (ICLR 2022 · similarity 0.91)
- Scaling Spherical CNNs (ICML 2023 · similarity 0.90)
- Spin-Weighted Spherical CNNs (NeurIPS 2020 · similarity 0.90)
- Scalable and Equivariant Spherical CNNs by Discrete-Continuous (DISCO) Convolutions (ICLR 2023 · similarity 0.90)
- Attention on the Sphere (NeurIPS 2025 · 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.