DW
Denny Wu
cs.LGstat.MLmath.STstat.THcs.AImath.PRcs.CVcs.ITmath.ITmath.OC
On Valency
published · living versionsW_q42y79mc·v1 · currentpublished
Gradient Descent on Two ReLU Neurons: Global Landscape and Bifurcation Dynamics
with Binghua Li, Mengzhe Li, Tianhao Wang
1 version
Preprints & journals
26 papers in the corpus · 2018–2026Sharp Capacity Scaling of Spectral Optimizers in Learning Associative Memory2603.26554v2 · Juno Kim, Eshaan Nichani, Denny Wu et al.2026 · 0 citationsarXiv
Learning to Recall with Transformers Beyond Orthogonal Embeddings.42583526 · Vural, Nuri Mert, Bietti, Alberto, Soltanolkotabi, Mahdi et al.2026 · 0 citations... International Conference on Learning Representations. 2026;2026:75057-75129
Full-Batch Gradient Descent Outperforms One-Pass SGD: Sample Complexity Separation in Single-Index Learning.42819568 · Kovačević, Filip, Ji, Hong Chang, Wu, Denny et al.2026 · 0 citationsProceedings of machine learning research. 2026;306:60661-60725
Understanding the Mechanisms of Fast Hyperparameter Transfer2512.22768v1 · Nikhil Ghosh, Denny Wu, Alberto Bietti2025 · 0 citationsarXiv
Learning Compositional Functions with Transformers from Easy-to-Hard Data2505.23683v1 · Zixuan Wang, Eshaan Nichani, Alberto Bietti et al.2025 · 0 citationsarXiv
Emergence and scaling laws in SGD learning of shallow neural networks2504.19983v2 · Yunwei Ren, Eshaan Nichani, Denny Wu et al.2025 · 0 citationsarXiv
Propagation of Chaos in One-hidden-layer Neural Networks beyond Logarithmic Time2504.13110v3 · Margalit Glasgow, Denny Wu, Joan Bruna2025 · 0 citationsarXiv
When Do Transformers Outperform Feedforward and Recurrent Networks? A Statistical Perspective2503.11272v1 · Alireza Mousavi-Hosseini, Clayton Sanford, Denny Wu et al.2025 · 0 citationsarXiv
Metastable Dynamics of Chain-of-Thought Reasoning: Provable Benefits of Search, RL and Distillation2502.01694v2 · Juno Kim, Denny Wu, Jason Lee et al.2025 · 0 citationsarXiv
Pretrained transformer efficiently learns low-dimensional target functions in-context2411.02544v1 · Kazusato Oko, Yujin Song, Taiji Suzuki et al.2024 · 1 citationarXiv
Learning Multi-Index Models with Neural Networks via Mean-Field Langevin Dynamics2408.07254v2 · Alireza Mousavi-Hosseini, Denny Wu, Murat A. Erdogdu2024 · 0 citationsarXiv
Learning sum of diverse features: computational hardness and efficient gradient-based training for ridge combinations2406.11828v1 · Kazusato Oko, Yujin Song, Taiji Suzuki et al.2024 · 0 citationsarXiv
Nonlinear spiked covariance matrices and signal propagation in deep neural networks2402.10127v1 · Zhichao Wang, Denny Wu, Zhou Fan2024 · 1 citationarXiv
Gradient-Based Feature Learning under Structured Data2309.03843v1 · Alireza Mousavi-Hosseini, Denny Wu, Taiji Suzuki et al.2023 · 1 citationarXiv
Convergence of mean-field Langevin dynamics: Time and space discretization, stochastic gradient, and variance reduction2306.07221v1 · Taiji Suzuki, Denny Wu, Atsushi Nitanda2023 · 0 citationsarXiv
Primal and Dual Analysis of Entropic Fictitious Play for Finite-sum Problems2303.02957v1 · Atsushi Nitanda, Kazusato Oko, Denny Wu et al.2023 · 0 citationsarXiv
High-dimensional Asymptotics of Feature Learning: How One Gradient Step Improves the Representation2205.01445v1 · Jimmy Ba, Murat A. Erdogdu, Taiji Suzuki et al.2022 · 17 citationsarXiv
Convex Analysis of the Mean Field Langevin Dynamics2201.10469v2 · Atsushi Nitanda, Denny Wu, Taiji Suzuki2022 · 3 citationsarXiv
Particle Dual Averaging: Optimization of Mean Field Neural Networks with Global Convergence Rate Analysis2012.15477v4 · Atsushi Nitanda, Denny Wu, Taiji Suzuki2020 · 2 citationsarXiv
When Does Preconditioning Help or Hurt Generalization?2006.10732v4 · Shun-ichi Amari, Jimmy Ba, Roger Grosse et al.2020 · 10 citationsarXiv
On the Optimal Weighted Regularization in Overparameterized Linear Regression2006.05800v4 · Denny Wu, Ji Xu2020 · 16 citationsarXiv
Stochastic Runge-Kutta Accelerates Langevin Monte Carlo and Beyond1906.07868v3 · Xuechen Li, Denny Wu, Lester Mackey et al.2019 · 25 citationsarXiv
Modeling the Biological Pathology Continuum with HSIC-regularized Wasserstein Auto-encoders1901.06618v1 · Denny Wu, Hirofumi Kobayashi, Charles Ding et al.2019 · 1 citationarXiv
Post Selection Inference with Incomplete Maximum Mean Discrepancy Estimator1802.06226v1 · Makoto Yamada, Denny Wu, Yao-Hung Hubert Tsai et al.2018 · 7 citationsarXiv
"Dependency Bottleneck" in Auto-encoding Architectures: an Empirical Study1802.05408v1 · Denny Wu, Yixiu Zhao, Yao-Hung Hubert Tsai et al.2018 · 12 citationsarXiv
Selecting the Best in GANs Family: a Post Selection Inference Framework1802.05411v2 · Yao-Hung Hubert Tsai, Makoto Yamada, Denny Wu et al.2018 · 2 citationsarXiv
Career total: 38 works. 26 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.