Research map / Learning theory and algorithms / Sample complexity
Rational neural networks
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 |
|---|---|---|---|
| replace ReLU activation with trainable rational activation (3,2) | 4-layer 50-node identification network · KdV PDE interpolation | validation MSE: 1.9\times 10^{-4} -> 1.2\times 10^{-7} | Helped |
| replace sinusoid activation with rational activation | identification network · KdV PDE interpolation | validation MSE: 3.3\times 10^{-6} -> 1.2\times 10^{-7} | Helped |
| replace polynomial (degree 3) activation with rational activation | identification network · KdV PDE interpolation | validation MSE: 3.6\times 10^{-5} -> 1.2\times 10^{-7} | Helped |
| replace ReLU/LeakyReLU with rational activations in generator and discriminator | ACGAN · MNIST generation | sample quality (qualitative) | Mixed |
Closest papers by meaning
- Rational Neural Networks have Expressivity Advantages (ICML 2026 · similarity 0.91)
- New Complexity-Theoretic Frontiers of Tractability for Neural Network Training (NeurIPS 2023 · similarity 0.88)
- On the Expressiveness of Rational ReLU Neural Networks With Bounded Depth (ICLR 2025 · similarity 0.87)
- Complexity of Neural Network Training and ETR: Extensions with Effectively Continuous Functions (AAAI 2024 · similarity 0.85)
- Better NTK Conditioning: A Free Lunch from (ReLU) Nonlinear Activation in Wide Neural Networks (NeurIPS 2025 · similarity 0.84)
- ReLU Network with Width $d+\mathcal{O}(1)$ Can Achieve Optimal Approximation Rate (ICML 2024 · similarity 0.84)
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.