Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Di He

Di He

Adversarial Noises Are Linearly Separable for (Nearly) Random Neural Networks

Jun 09, 2022
Huishuai Zhang, Da Yu, Yiping Lu, Di He

Figure 1 for Adversarial Noises Are Linearly Separable for (Nearly) Random Neural Networks

Figure 2 for Adversarial Noises Are Linearly Separable for (Nearly) Random Neural Networks

* 13 pages

Access Paper or Ask Questions

Is $L^2$ Physics-Informed Loss Always Suitable for Training Physics-Informed Neural Network?

Jun 04, 2022
Chuwei Wang, Shanda Li, Di He, Liwei Wang

Figure 1 for Is $L^2$ Physics-Informed Loss Always Suitable for Training Physics-Informed Neural Network?

Figure 2 for Is $L^2$ Physics-Informed Loss Always Suitable for Training Physics-Informed Neural Network?

Figure 3 for Is $L^2$ Physics-Informed Loss Always Suitable for Training Physics-Informed Neural Network?

Figure 4 for Is $L^2$ Physics-Informed Loss Always Suitable for Training Physics-Informed Neural Network?

Access Paper or Ask Questions

Your Transformer May Not be as Powerful as You Expect

May 26, 2022
Shengjie Luo, Shanda Li, Shuxin Zheng, Tie-Yan Liu, Liwei Wang, Di He

Figure 1 for Your Transformer May Not be as Powerful as You Expect

Figure 2 for Your Transformer May Not be as Powerful as You Expect

Figure 3 for Your Transformer May Not be as Powerful as You Expect

Figure 4 for Your Transformer May Not be as Powerful as You Expect

* Preprint. Work in Progress

Access Paper or Ask Questions

METRO: Efficient Denoising Pretraining of Large Scale Autoencoding Language Models with Model Generated Signals

Apr 16, 2022
Payal Bajaj, Chenyan Xiong, Guolin Ke, Xiaodong Liu, Di He, Saurabh Tiwary, Tie-Yan Liu, Paul Bennett, Xia Song, Jianfeng Gao

Figure 1 for METRO: Efficient Denoising Pretraining of Large Scale Autoencoding Language Models with Model Generated Signals

Figure 2 for METRO: Efficient Denoising Pretraining of Large Scale Autoencoding Language Models with Model Generated Signals

Figure 3 for METRO: Efficient Denoising Pretraining of Large Scale Autoencoding Language Models with Model Generated Signals

Figure 4 for METRO: Efficient Denoising Pretraining of Large Scale Autoencoding Language Models with Model Generated Signals

* Update details in scaled initialization and add acknowledgement

Access Paper or Ask Questions

An Empirical Study of Graphormer on Large-Scale Molecular Modeling Datasets

Mar 14, 2022
Yu Shi, Shuxin Zheng, Guolin Ke, Yifei Shen, Jiacheng You, Jiyan He, Shengjie Luo, Chang Liu, Di He, Tie-Yan Liu

Figure 1 for An Empirical Study of Graphormer on Large-Scale Molecular Modeling Datasets

Figure 2 for An Empirical Study of Graphormer on Large-Scale Molecular Modeling Datasets

Figure 3 for An Empirical Study of Graphormer on Large-Scale Molecular Modeling Datasets

Figure 4 for An Empirical Study of Graphormer on Large-Scale Molecular Modeling Datasets

* Wrong dual-submission (arXiv:2203.04810) with negligently

Access Paper or Ask Questions