works on

From the 1 of 6 linked papers with an AI index.

activity
20182021
most citedTowards Resolving the Implicit Bias of Gradient Descent for Matrix Factorization: Greedy Low-Rank Learning

13 citations · 25 across the 3 of their papers we have counts for

collaborators

6 papers

cs.LG20212 cited

Gradient Descent on Two-layer Nets: Margin Maximization and Simplicity Bias

Kaifeng Lyu, Zhiyuan Li, Runzhe Wang +1

The generalization mystery of overparametrized deep nets has motivated efforts to understand how gradient descent (GD) converges to low-loss solutions that generalize well. Real-li…

cs.LG202013 cited

Towards Resolving the Implicit Bias of Gradient Descent for Matrix Factorization: Greedy Low-Rank Learning

Zhiyuan Li, Yuping Luo, Kaifeng Lyu

Matrix factorization is a simple and natural test-bed to investigate the implicit regularization of gradient descent. Gunasekar et al. (2017) conjectured that Gradient Flow with in…

cs.LG202010 cited

Reconciling Modern Deep Learning with Traditional Optimization Analyses: The Intrinsic Learning Rate

Zhiyuan Li, Kaifeng Lyu, Sanjeev Arora

The paper proposes an 'intrinsic learning rate' that combines the usual learning rate and weight decay to model the training dynamics of normalized deep networks via stochastic dif…

cs.LG2019

Gradient Descent Maximizes the Margin of Homogeneous Neural Networks

Kaifeng Lyu, Jian Li

In this paper, we study the implicit regularization of the gradient descent algorithm in homogeneous neural networks, including fully-connected and convolutional neural networks wi…

cs.LG2018

Theoretical Analysis of Auto Rate-Tuning by Batch Normalization

Sanjeev Arora, Zhiyuan Li, Kaifeng Lyu

Batch Normalization (BN) has become a cornerstone of deep learning across diverse architectures, appearing to help optimization as well as generalization. While the idea makes intu…

cs.DS2018

Single-Source Bottleneck Path Algorithm Faster than Sorting for Sparse Graphs

Ran Duan, Kaifeng Lyu, Hongxun Wu +1

In a directed graph with a capacity on every edge, a \emph{bottleneck path} (or \emph{widest path}) between two vertices is a path maximizing the minimum capacity of edge…