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

Add to Chrome

Add to Firefox

Picture for Bang An

Bang An

HintNet: Hierarchical Knowledge Transfer Networks for Traffic Accident Forecasting on Heterogeneous Spatio-Temporal Data

Mar 07, 2022
Bang An, Amin Vahedian, Xun Zhou, W. Nick Street, Yanhua Li

Figure 1 for HintNet: Hierarchical Knowledge Transfer Networks for Traffic Accident Forecasting on Heterogeneous Spatio-Temporal Data

Figure 2 for HintNet: Hierarchical Knowledge Transfer Networks for Traffic Accident Forecasting on Heterogeneous Spatio-Temporal Data

Figure 3 for HintNet: Hierarchical Knowledge Transfer Networks for Traffic Accident Forecasting on Heterogeneous Spatio-Temporal Data

Figure 4 for HintNet: Hierarchical Knowledge Transfer Networks for Traffic Accident Forecasting on Heterogeneous Spatio-Temporal Data

* 9 pages, 2022 SIAM International Conference on Data Mining

Access Paper or Ask Questions

Understanding the Generalization Benefit of Model Invariance from a Data Perspective

Nov 10, 2021
Sicheng Zhu, Bang An, Furong Huang

Figure 1 for Understanding the Generalization Benefit of Model Invariance from a Data Perspective

Figure 2 for Understanding the Generalization Benefit of Model Invariance from a Data Perspective

Figure 3 for Understanding the Generalization Benefit of Model Invariance from a Data Perspective

Figure 4 for Understanding the Generalization Benefit of Model Invariance from a Data Perspective

* Accepted to NeurIPS 2021

Access Paper or Ask Questions

Repulsive Attention: Rethinking Multi-head Attention as Bayesian Inference

Sep 20, 2020
Bang An, Jie Lyu, Zhenyi Wang, Chunyuan Li, Changwei Hu, Fei Tan, Ruiyi Zhang, Yifan Hu, Changyou Chen

Figure 1 for Repulsive Attention: Rethinking Multi-head Attention as Bayesian Inference

Figure 2 for Repulsive Attention: Rethinking Multi-head Attention as Bayesian Inference

Figure 3 for Repulsive Attention: Rethinking Multi-head Attention as Bayesian Inference

Figure 4 for Repulsive Attention: Rethinking Multi-head Attention as Bayesian Inference

* accepted by EMNLP2020

Access Paper or Ask Questions

Towards Faithful Neural Table-to-Text Generation with Content-Matching Constraints

May 03, 2020
Zhenyi Wang, Xiaoyang Wang, Bang An, Dong Yu, Changyou Chen

Figure 1 for Towards Faithful Neural Table-to-Text Generation with Content-Matching Constraints

Figure 2 for Towards Faithful Neural Table-to-Text Generation with Content-Matching Constraints

Figure 3 for Towards Faithful Neural Table-to-Text Generation with Content-Matching Constraints

Figure 4 for Towards Faithful Neural Table-to-Text Generation with Content-Matching Constraints

* Accepted at ACL2020

Access Paper or Ask Questions

Guess First to Enable Better Compression and Adversarial Robustness

Jan 10, 2020
Sicheng Zhu, Bang An, Shiyu Niu

Figure 1 for Guess First to Enable Better Compression and Adversarial Robustness

Figure 2 for Guess First to Enable Better Compression and Adversarial Robustness

Figure 3 for Guess First to Enable Better Compression and Adversarial Robustness

Figure 4 for Guess First to Enable Better Compression and Adversarial Robustness

* Accepted by NeurIPS 2019 workshop on Information Theory and Machine Learning

Access Paper or Ask Questions