Alert button
Picture for Huishuai Zhang

Huishuai Zhang

Alert button

Adversarial Training with Rectified Rejection

May 31, 2021
Tianyu Pang, Huishuai Zhang, Di He, Yinpeng Dong, Hang Su, Wei Chen, Jun Zhu, Tie-Yan Liu

Figure 1 for Adversarial Training with Rectified Rejection
Figure 2 for Adversarial Training with Rectified Rejection
Figure 3 for Adversarial Training with Rectified Rejection
Figure 4 for Adversarial Training with Rectified Rejection
Viaarxiv icon

Do Not Let Privacy Overbill Utility: Gradient Embedding Perturbation for Private Learning

Feb 26, 2021
Da Yu, Huishuai Zhang, Wei Chen, Tie-Yan Liu

Figure 1 for Do Not Let Privacy Overbill Utility: Gradient Embedding Perturbation for Private Learning
Figure 2 for Do Not Let Privacy Overbill Utility: Gradient Embedding Perturbation for Private Learning
Figure 3 for Do Not Let Privacy Overbill Utility: Gradient Embedding Perturbation for Private Learning
Figure 4 for Do Not Let Privacy Overbill Utility: Gradient Embedding Perturbation for Private Learning
Viaarxiv icon

BN-invariant sharpness regularizes the training model to better generalization

Jan 08, 2021
Mingyang Yi, Huishuai Zhang, Wei Chen, Zhi-Ming Ma, Tie-Yan Liu

Figure 1 for BN-invariant sharpness regularizes the training model to better generalization
Figure 2 for BN-invariant sharpness regularizes the training model to better generalization
Figure 3 for BN-invariant sharpness regularizes the training model to better generalization
Figure 4 for BN-invariant sharpness regularizes the training model to better generalization
Viaarxiv icon

Well-Conditioned Methods for Ill-Conditioned Systems: Linear Regression with Semi-Random Noise

Aug 04, 2020
Jerry Li, Aaron Sidford, Kevin Tian, Huishuai Zhang

Figure 1 for Well-Conditioned Methods for Ill-Conditioned Systems: Linear Regression with Semi-Random Noise
Figure 2 for Well-Conditioned Methods for Ill-Conditioned Systems: Linear Regression with Semi-Random Noise
Figure 3 for Well-Conditioned Methods for Ill-Conditioned Systems: Linear Regression with Semi-Random Noise
Figure 4 for Well-Conditioned Methods for Ill-Conditioned Systems: Linear Regression with Semi-Random Noise
Viaarxiv icon

Membership Inference with Privately Augmented Data Endorses the Benign while Suppresses the Adversary

Jul 21, 2020
Da Yu, Huishuai Zhang, Wei Chen, Jian Yin, Tie-Yan Liu

Figure 1 for Membership Inference with Privately Augmented Data Endorses the Benign while Suppresses the Adversary
Figure 2 for Membership Inference with Privately Augmented Data Endorses the Benign while Suppresses the Adversary
Figure 3 for Membership Inference with Privately Augmented Data Endorses the Benign while Suppresses the Adversary
Figure 4 for Membership Inference with Privately Augmented Data Endorses the Benign while Suppresses the Adversary
Viaarxiv icon

Adai: Separating the Effects of Adaptive Learning Rate and Momentum Inertia

Jul 17, 2020
Zeke Xie, Xinrui Wang, Huishuai Zhang, Issei Sato, Masashi Sugiyama

Figure 1 for Adai: Separating the Effects of Adaptive Learning Rate and Momentum Inertia
Figure 2 for Adai: Separating the Effects of Adaptive Learning Rate and Momentum Inertia
Figure 3 for Adai: Separating the Effects of Adaptive Learning Rate and Momentum Inertia
Figure 4 for Adai: Separating the Effects of Adaptive Learning Rate and Momentum Inertia
Viaarxiv icon

On Layer Normalization in the Transformer Architecture

Feb 12, 2020
Ruibin Xiong, Yunchang Yang, Di He, Kai Zheng, Shuxin Zheng, Chen Xing, Huishuai Zhang, Yanyan Lan, Liwei Wang, Tie-Yan Liu

Figure 1 for On Layer Normalization in the Transformer Architecture
Figure 2 for On Layer Normalization in the Transformer Architecture
Figure 3 for On Layer Normalization in the Transformer Architecture
Figure 4 for On Layer Normalization in the Transformer Architecture
Viaarxiv icon

Gradient Perturbation is Underrated for Differentially Private Convex Optimization

Nov 26, 2019
Da Yu, Huishuai Zhang, Wei Chen, Tie-Yan Liu, Jian Yin

Figure 1 for Gradient Perturbation is Underrated for Differentially Private Convex Optimization
Figure 2 for Gradient Perturbation is Underrated for Differentially Private Convex Optimization
Figure 3 for Gradient Perturbation is Underrated for Differentially Private Convex Optimization
Figure 4 for Gradient Perturbation is Underrated for Differentially Private Convex Optimization
Viaarxiv icon