Research map / Architectures and efficiency / Neural architecture search
A Classification of $G$-invariant Shallow Neural Networks
Devanshu Agrawal, James Ostrowski
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 |
|---|---|---|---|
| allow type 2 (signed permutation) architectures beyond ordinary permutation-representation (type 1) ones | G-invariant shallow neural network architectures · D4 on four elements | number of architectures: five type 1 vs seven type 2 | Improved |
| use signed instead of unsigned permutation representation for C6 hidden layer | C6-invariant SNN · C6 permutation representation | topology of architecture | Mixed |
| let bias go to zero in the type 1 architecture | C6 rotation type 1 SNN · 2D rotation representation of C6 | function-space distance to type 2 | Mixed |
Closest papers by meaning
- Densely Connected G-invariant Deep Neural Networks with Signed Permutation Representations (JMLR 2023 · similarity 0.90)
- On the Universality of Invariant Networks (ICML 2019 · similarity 0.87)
- G-SGD: Optimizing ReLU Neural Networks in its Positively Scale-Invariant Space (ICLR 2019 · similarity 0.86)
- Topological obstruction to the training of shallow ReLU neural networks (NeurIPS 2024 · similarity 0.85)
- The Selective $G$-Bispectrum and its Inversion: Applications to $G$-Invariant Networks (NeurIPS 2024 · similarity 0.85)
- Monomial Matrix Group Equivariant Neural Functional Networks (NeurIPS 2024 · 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.
