Research map / Optimization / Optimizers
Nonlinear Sufficient Dimension Reduction with a Stochastic Neural Network
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 StoNet instead of GSIR for nonlinear SDR | Binary classification SDR · thyroid (q=1) | misclassification rate: 0.0687(.0068) vs 0.0640(.0063) | Hurt |
| Use StoNet instead of GSAVE for nonlinear SDR | Binary classification SDR · german (q=5) | misclassification rate: 0.2368(.0050) vs 0.2323(.0050) | Hurt |
| Use StoNet instead of GSAVE for nonlinear SDR | Binary classification SDR · waveform (q=5) | misclassification rate: 0.1091(.0010) vs 0.1095(.0016) | No clear effect |
| Use StoNet for dimension reduction rather than GSAVE | MNIST sub-training set classifier · MNIST sub-training set (q=392) | misclassification rate: 0.0456 vs 0.0535 | Helped |
| Use StoNet for dimension reduction rather than GSIR | MNIST sub-training set classifier · MNIST sub-training set (q=10) | misclassification rate: 0.0825 vs 0.0872 | Helped |
| Use StoNet rather than GSIR | MNIST sub-training set SDR · MNIST sub-training set | Average Time(s): 96.18 vs 16005.59 | Helped |
| Use StoNet rather than PCA for dimension reduction | CT slice regression model · Relative location of CT slices (q=192) | MSE: 0.0027(.0000) -> 0.0002(.0000) | Helped |
| Use StoNet rather than PCA for dimension reduction | CT slice regression model · Relative location of CT slices (q=3) | Pearson correlation: 0.7858(.0004) -> 0.9965(.0002) | Helped |
| Fit an unconstrained complicated neural network for the regression function rather than StoNet's structured stochastic network | Deep SDR · M1 simulated example | p-value for dependence test: 0.012 vs 0.4068 | Hurt |
Closest papers by meaning
- Neural Networks Perform Sufficient Dimension Reduction (AAAI 2025 · similarity 0.86)
- Deep Principal Support Vector Machines for Nonlinear Sufficient Dimension Reduction (ICML 2025 · similarity 0.86)
- The Power of Random Features and the Limits of Distribution-Free Gradient Descent (ICML 2025 · similarity 0.85)
- Enhancing Sufficient Dimension Reduction via Hellinger Correlation (ICML 2024 · similarity 0.85)
- A Study of Bayesian Neural Network Surrogates for Bayesian Optimization (ICLR 2024 · similarity 0.84)
- Parameter Expanded Stochastic Gradient Markov Chain Monte Carlo (ICLR 2025 · 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.