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

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Dinghuai Zhang

Dinghuai Zhang

Invariance Principle Meets Information Bottleneck for Out-of-Distribution Generalization


Jun 11, 2021
Kartik Ahuja, Ethan Caballero, Dinghuai Zhang, Yoshua Bengio, Ioannis Mitliagkas, Irina Rish

Figure 1 for Invariance Principle Meets Information Bottleneck for Out-of-Distribution Generalization
Figure 2 for Invariance Principle Meets Information Bottleneck for Out-of-Distribution Generalization
Figure 3 for Invariance Principle Meets Information Bottleneck for Out-of-Distribution Generalization
Figure 4 for Invariance Principle Meets Information Bottleneck for Out-of-Distribution Generalization

  Access Paper or Ask Questions

Can Subnetwork Structure be the Key to Out-of-Distribution Generalization?


Jun 05, 2021
Dinghuai Zhang, Kartik Ahuja, Yilun Xu, Yisen Wang, Aaron Courville

Figure 1 for Can Subnetwork Structure be the Key to Out-of-Distribution Generalization?
Figure 2 for Can Subnetwork Structure be the Key to Out-of-Distribution Generalization?
Figure 3 for Can Subnetwork Structure be the Key to Out-of-Distribution Generalization?
Figure 4 for Can Subnetwork Structure be the Key to Out-of-Distribution Generalization?
* Accepted to ICML2021 as long talk 
  Access Paper or Ask Questions

Neural Approximate Sufficient Statistics for Implicit Models


Oct 20, 2020
Yanzhi Chen, Dinghuai Zhang, Michael Gutmann, Aaron Courville, Zhanxing Zhu

Figure 1 for Neural Approximate Sufficient Statistics for Implicit Models
Figure 2 for Neural Approximate Sufficient Statistics for Implicit Models
Figure 3 for Neural Approximate Sufficient Statistics for Implicit Models
Figure 4 for Neural Approximate Sufficient Statistics for Implicit Models

  Access Paper or Ask Questions

Informative Dropout for Robust Representation Learning: A Shape-bias Perspective


Aug 10, 2020
Baifeng Shi, Dinghuai Zhang, Qi Dai, Zhanxing Zhu, Yadong Mu, Jingdong Wang

* Accepted to ICML2020. Code is available at https://github.com/bfshi/InfoDrop 
  Access Paper or Ask Questions

Black-Box Certification with Randomized Smoothing: A Functional Optimization Based Framework


Feb 21, 2020
Dinghuai Zhang, Mao Ye, Chengyue Gong, Zhanxing Zhu, Qiang Liu

Figure 1 for Black-Box Certification with Randomized Smoothing: A Functional Optimization Based Framework
Figure 2 for Black-Box Certification with Randomized Smoothing: A Functional Optimization Based Framework
Figure 3 for Black-Box Certification with Randomized Smoothing: A Functional Optimization Based Framework
Figure 4 for Black-Box Certification with Randomized Smoothing: A Functional Optimization Based Framework

  Access Paper or Ask Questions

You Only Propagate Once: Accelerating Adversarial Training via Maximal Principle


May 23, 2019
Dinghuai Zhang, Tianyuan Zhang, Yiping Lu, Zhanxing Zhu, Bin Dong

Figure 1 for You Only Propagate Once: Accelerating Adversarial Training via Maximal Principle
Figure 2 for You Only Propagate Once: Accelerating Adversarial Training via Maximal Principle
Figure 3 for You Only Propagate Once: Accelerating Adversarial Training via Maximal Principle
Figure 4 for You Only Propagate Once: Accelerating Adversarial Training via Maximal Principle

  Access Paper or Ask Questions