Research map / Architectures and efficiency / RNNs and normalization
Efficient Algorithms for Learning Depth-2 Neural Networks with General ReLU Activations
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 multiple consecutive higher-order Hermite tensors instead of relying on a single tensor | Proposed algorithm for biased depth-2 ReLU networks · Gaussian inputs, W^{⊙ℓ} linearly independent | parameter recovery | Helped |
| Approximate the contribution of large-magnitude-bias units by a linear function and fit the residual with truncated linear regression | Proposed algorithm for biased depth-2 ReLU networks · Gaussian inputs, including units with large-magnitude biases | L2 error: ≤ ε² | Helped |
| Use tensor decompositions of Hermite coefficients to recover parameters up to sign | Proposed algorithm for biased depth-2 ReLU networks · Gaussian inputs, W^{⊙ℓ} linearly independent | parameter recovery | Helped |
| Recover all parameters without restricting bias magnitude | Proposed algorithm for biased depth-2 ReLU networks · Gaussian inputs, large-magnitude biases | parameter recovery | Hurt |
Closest papers by meaning
- An Exact Poly-Time Membership-Queries Algorithm for Extracting a Three-Layer ReLU Network (ICLR 2023 · similarity 0.88)
- Compelling ReLU Networks to Exhibit Exponentially Many Linear Regions at Initialization and During Training (ICML 2025 · similarity 0.88)
- Neural Networks with Small Weights and Depth-Separation Barriers (NeurIPS 2020 · similarity 0.88)
- ReLUs Are Sufficient for Learning Implicit Neural Representations (ICML 2024 · similarity 0.87)
- Optimal Rates for Generalization of Gradient Descent for Deep ReLU Classification (NeurIPS 2025 · similarity 0.86)
- Neural Characteristic Activation Analysis and Geometric Parameterization for ReLU Networks (NeurIPS 2024 · 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.