Research map / Learning theory and algorithms / Sample complexity
Smoothing the Landscape Boosts the Signal for SGD: Optimal Sample Complexity for Learning Single Index Models
Alex Damian, Eshaan Nichani, Rong Ge, Jason D. Lee
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 |
|---|---|---|---|
| Smooth the loss landscape during online SGD | Online SGD for single-index models · Single-index models with information exponent k* | sample complexity: d^{k^\star-1} -> d^{k^\star/2} | Improved |
| Use λ = O(1), equivalent to zero smoothing (ablation) | Smoothed online SGD · Single-index models with information exponent k* | sample complexity: d^{k^\star/2} -> d^{k^\star-1} | Hurt |
| Set smoothing to λ = d^{1/4} (ablation) | Smoothed online SGD · Single-index models with information exponent k* | sample complexity: d^{k^\star-1} -> d^{k^\star/2} | Improved |
| Smooth the loss landscape | Online SGD for single-index models · Initialization region, α ≲ λ d^{-1/2} | gradient signal-to-noise ratio | Improved |
Closest papers by meaning
- Learning Orthogonal Multi-Index Models: A Fine-Grained Information Exponent Analysis (NeurIPS 2025 · similarity 0.90)
- Learning single index models via harmonic decomposition (NeurIPS 2025 · similarity 0.88)
- Sample and Computationally Efficient Robust Learning of Gaussian Single-Index Models (NeurIPS 2024 · similarity 0.87)
- Neural network learns low-dimensional polynomials with SGD near the information-theoretic limit (NeurIPS 2024 · similarity 0.87)
- From Information to Generative Exponent: Learning Rate Induces Phase Transitions in SGD (NeurIPS 2025 · similarity 0.87)
- The Generative Leap: Tight Sample Complexity for Efficiently Learning Gaussian Multi-Index Models (NeurIPS 2025 · similarity 0.87)
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.
