Research map / Probabilistic and scientific modeling / Dynamical systems
Numerical Analysis near Singularities in RBF Networks
Weili Guo, Haikun Wei, Yew-Soon Ong, Jaime Rubio Hervas, Junsheng Zhao, Hai Wang, Kanjian Zhang
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 |
|---|---|---|---|
| compare elimination vs overlap singularities as a cause of disrupted learning in 2-unit RBF networks | RBF network k=2 · 2-hidden-unit RBF, ALE simulation | percent of runs affected: 7.5 percent overlap vs 12.5 percent elimination | Mixed |
| increase hidden units from 2 to 10 | RBF network · RBF ALE simulation | percent of runs unaffected by singularities: nearly 68 percent -> 67.5 percent | No clear effect |
| move from low-dimensional synthetic data to the 7-dimensional ECSSD features | RBF network · ECSSD vs synthetic | overlap singularity runs: 109 -> 3 | Improved |
| move from low-dimensional synthetic data to ECSSD | RBF network · ECSSD | output-weight-near-0 runs: 12.3 percent -> only 1 percent | Improved |
| high-dimensional real data raises the share of elimination-singularity runs | RBF network k=90 · ECSSD | percent of runs affected by elimination singularity: 21 percent | Hurt |
| student centers initialised far from teacher centers | RBF network k=2 · 2-hidden-unit RBF, ALE simulation | runs ending with output weight near 0: nearly 12 percent | Hurt |
Closest papers by meaning
- The Convergence Rate of Neural Networks for Learned Functions of Different Frequencies (NeurIPS 2019 · similarity 0.84)
- Universal Sharpness Dynamics in Neural Network Training: Fixed Point Analysis, Edge of Stability, and Route to Chaos (ICLR 2025 · similarity 0.84)
- Alternating Gradient Flows: A Theory of Feature Learning in Two-layer Neural Networks (NeurIPS 2025 · similarity 0.83)
- Asymptotics of feature learning in two-layer networks after one gradient-step (ICML 2024 · similarity 0.82)
- Neural Collapse under Gradient Flow on Shallow ReLU Networks for Orthogonally Separable Data (NeurIPS 2025 · similarity 0.82)
- Learning quadratic neural networks in high dimensions: SGD dynamics and scaling laws (NeurIPS 2025 · similarity 0.82)
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.
