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

256 citations · 487 across the 10 of their papers we have counts for

collaborators

10 papers

cs.LG20204 cited

Provable Representation Learning for Imitation Learning via Bi-level Optimization

Sanjeev Arora, Simon S. Du, Sham Kakade +2

A common strategy in modern learning systems is to learn a representation that is useful for many tasks, a.k.a. representation learning. We study this strategy in the imitation lea…

cs.LG20207 cited

Over-parameterized Adversarial Training: An Analysis Overcoming the Curse of Dimensionality

Yi Zhang, Orestis Plevrakis, Simon S. Du +3

Adversarial training is a popular method to give neural nets robustness against adversarial perturbations. In practice adversarial training leads to low robust training loss. Howev…

cs.LG202023 cited

Agnostic Q-learning with Function Approximation in Deterministic Systems: Tight Bounds on Approximation Error and Sample Complexity

Simon S. Du, Jason D. Lee, Gaurav Mahajan +1

The current paper studies the problem of agnostic -learning with function approximation in deterministic systems where the optimal -function is approximable by a function in…

stat.ML201953 cited

Optimism in Reinforcement Learning with Generalized Linear Function Approximation

Yining Wang, Ruosong Wang, Simon S. Du +1

We design a new provably efficient algorithm for episodic reinforcement learning with generalized linear function approximation. We analyze the algorithm under a new expressivity a…

cs.LG201977 cited

Enhanced Convolutional Neural Tangent Kernels

Zhiyuan Li, Ruosong Wang, Dingli Yu +4

Recent research shows that for training with loss, convolutional neural networks (CNNs) whose width (number of channels in convolutional layers) goes to infinity correspon…

cs.LG201922 cited

Towards Understanding the Importance of Shortcut Connections in Residual Networks

Tianyi Liu, Minshuo Chen, Mo Zhou +3

Residual Network (ResNet) is undoubtedly a milestone in deep learning. ResNet is equipped with shortcut connections between layers, and exhibits efficient training using simple fir…