Simon S. Du

Gradient Descent Provably Optimizes Over-parameterized Neural Networks

Simon S. Du, Xiyu Zhai, Barnabás Póczos et al. · arXiv (Cornell University) · 2018 · 418 citations · Full text

Model Optimization, Hyperparameter Estimation, Engineering +12

Gradient descent finds global minima of deep neural networks

Simon S. Du, Jason D. Lee, Haochuan Li et al. · 2019 · 417 citations

Artificial Intelligence, Model Optimization, Convolutional Neural Network +11

On Exact Computation with an Infinitely Wide Neural Net

Sanjeev Arora, Simon S. Du, Wei Hu et al. · arXiv (Cornell University) · 2019 · 395 citations · Full text

Artificial Intelligence, Convolutional Neural Network, Engineering +17

Fine-Grained Analysis of Optimization and Generalization for Overparameterized Two-Layer Neural Networks

Sanjeev Arora, Simon S. Du, Wei Hu et al. · arXiv (Cornell University) · 2019 · 259 citations · Full text

Artificial Intelligence, Convolutional Neural Network, Engineering +16

Fine-Grained Analysis of Optimization and Generalization for\n Overparameterized Two-Layer Neural Networks

Sanjeev Arora, Simon S. Du, Wei Hu et al. · arXiv (Cornell University) · 2019 · 237 citations · Full text