Mengnan Du

Ph.D. Student
Department of Computer Science & Engineering


I am currently a 4th year Ph.D. student in Computer Science at the CSE department of Texas A&M University. My advisor is Dr. Xia Hu.  My research is on interpretable machine learning, with a particular interest in the areas of DNN interpretability. I am also interested in areas of fairness in deep learning, adversarial detection, fake news detection, and medical diagnosis.

Please visit new webpage at:

Selected Publications (full list)

  • Mengnan Du, Ninghao Liu,  Xia Hu, "Techniques for Interpretable Machine Learning",  Communications of the ACM (CACM), Review Article, 2020. [PDF [CACM version]  [Video]  Highlighted Article on the Cover Page
  • Mengnan Du, Fan Yang, Na Zou, and Xia Hu, "Fairness in Deep Learning: A Computational Perspective", accepted by IEEE Intelligent Systems, 2020. [PDF]  [Github Awesome List
  • Mengnan Du*, Shiva Pentyala*, Yuening Li,  and Xia Hu, "Towards Generalizable Deepfake Detection with Locality-aware AutoEncoder", The 29th ACM International Conference on Information and Knowledge Management (CIKM), 2020. (*Equal contribution)  [PDF[Github]
  • Mengnan Du, Ninghao Liu, Fan Yang,  Xia Hu, "Learning Credible DNNs via Incorporating Prior Knowledge and Model Local Explanation", accepted by Knowledge and Information Systems (KAIS), 2020.
  • Ruixiang Tang, Mengnan Du, Ninghao Liu, Fan Yang and Xia Hu, "An Embarrassingly Simple Approach for Trojan Attack in Deep Neural Networks", The 26rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), 2020. [PDF[Code]
  • Haofan Wang, Zifan Wang, Mengnan Du, Fan Yang, Zijian Zhang, Sirui Ding, Piotr Mardziel, Xia Hu, "Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks", CVPR Workshop on Fair, Data-Efficient and Trusted Computer Vision, 2020.  [PDF[Code]
  • Mengnan Du, Ninghao Liu, Fan Yang,  Xia Hu, "Learning Credible Deep Neural Networks with Rationale Regularization",  IEEE International Conference on Data Mining (ICDM), 2019.  [PDF]  Best Paper Award Candidate
  • Mengnan Du, Ninghao Liu, Fan Yang, Shuiwang Ji,  Xia Hu, "On Attribution of Recurrent Neural Network Predictions via Additive Decomposition",  The Web Conference (WWW), 2019.  [PDF] [Code[Slides]  Best Paper Award Candidate
  • Fan Yang*, Shiva Pentyala*, Sina Mohseni*, Mengnan Du*, Hao Yuan*, Rhema Linder, Eric Ragan, Shuiwang Ji and Xia Hu, "XFake: Explainable Fake News Detector with Visualizations",  The Web Conference (WWW), demo track, 2019. (*Equal contribution)  [PDF
  • Ninghao Liu, Mengnan Du,  Xia Hu, "Representation Interpretation with Spatial Encoding and Multimodal Analytics",  ACM International Conference on Web Search and Data Mining (WSDM), 2019. [PDF]
  • Mengnan Du, Ninghao Liu, Qingquan Song, and Xia Hu, "Towards Explanation of DNN-based Prediction with Guided Feature Inversion", The 24rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), 2018. [PDF] [Code[Poster]

Preprint/Recent work
  • Ruixiang Tang, Mengnan DuYuening Li, Zirui Liu  and Xia Hu, "Mitigating Gender Bias in Captioning Systems", arXiv:2006.08315, 2020. [PDF]
  • Fan Yang, Mengnan Du, and Xia Hu, "Evaluating Explanation Without Ground Truth in Interpretable Machine Learning",  arXiv:1907.06831, 2019. [PDF
  • Ninghao LiuMengnan Du, and Xia Hu, "Adversarial Machine Learning: An Interpretation Perspective",  arXiv:2004.11488, 2020. [PDF

Industry Experience

          Project: Design object detection and speech recognition algorithm for an AR application.
  • Machine Learning Research Intern (05/2020 to 08/2020), Adobe, San Jose (Remotely in College Station, Texas)