Profiles2023-2026
Denny Wu
New York University
13Papers
4Last author
17Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers13
- Full-Batch Gradient Descent Outperforms One-Pass SGD: Sample Complexity Separation in Single-Index LearningAuthor 3 of 5ICML 2026 · 2 citations
- Learning to Recall with Transformers Beyond Orthogonal EmbeddingsAuthor 4 of 4ICLR 2026 · 1 citation
- Emergence and scaling laws in SGD learning of shallow neural networksAuthor 3 of 4NeurIPS 2025 · 33 citations
- Learning quadratic neural networks in high dimensions: SGD dynamics and scaling lawsAuthor 5 of 5NeurIPS 2025 · 23 citations
- When Do Transformers Outperform Feedforward and Recurrent Networks? A Statistical PerspectiveAuthor 3 of 4NeurIPS 2025 · 6 citations
- How Does Label Noise Gradient Descent Improve Generalization in the Low SNR Regime?Author 5 of 9NeurIPS 2025 · 4 citations
- Learning Multi-Index Models with Neural Networks via Mean-Field Langevin DynamicsAuthor 2 of 3ICLR 2025
- Neural network learns low-dimensional polynomials with SGD near the information-theoretic limitAuthor 4 of 4NeurIPS 2024 · 49 citations
- Pretrained Transformer Efficiently Learns Low-Dimensional Target Functions In-ContextAuthor 4 of 4NeurIPS 2024 · 34 citations
- Learning in the Presence of Low-dimensional Structure: A Spiked Random Matrix PerspectiveAuthor 5 of 6NeurIPS 2023 · 47 citations
- Gradient-Based Feature Learning under Structured DataAuthor 2 of 4NeurIPS 2023 · 36 citations
- Feature learning via mean-field Langevin dynamics: classifying sparse parities and beyondAuthor 2 of 4NeurIPS 2023 · 17 citations
- Mean-field Langevin dynamics: Time-space discretization, stochastic gradient, and variance reductionAuthor 2 of 3NeurIPS 2023 · 10 citations
