Shu Hu 胡暑

Shu Hu 

Ph.D.
Heinz College
Carnegie Mellon University
Email: shuhu [at] cmu (dot) edu
Google Scholar

About Me

I am a Post-Doctoral Fellow at the Heinz College of Carnegie Mellon University, where I am working with Prof. George H. Chen.

I received my Ph.D. degree in Computer Science and Engineering from University at Buffalo (UBuffalo), SUNY in 2022 under the advising of Prof. Siwei Lyu. I spend my first and second year of Ph.D. at University at Albany (UAlbany), SUNY. During the first year of my Ph.D., I was working with Prof. Feng Chen. I received my Master of Arts (Mathematics) degree from University at Albany, SUNY in 2020 under the advising of Prof. Yiming Ying and I also received my Master of Engineering degree from University of Science and Technology of China (USTC) in 2016 under the advising of Prof. En-Hong Chen. I was previously a visiting student (2014-2015) at University of South Australia (UniSA), where I worked with Prof. Jiuyong Li and Dr. Thuc Duy Le.

My current research interests include digital media forensics, computer vision, and machine learning, especially ranking and optimization problems in Machine Learning and Deep Learning. I also interest in information security and biomedical informatics.

News

  • (07/18/2022) Our survey is avaiable on ArXiv. Please check it [HERE]. This is the first comprehensive survey on Rank-based Decomposable Losses in Machine Learning. Most importantly, we introduce a new set function, aggregator, to formulate aggregate loss and individual loss. We hope this survey can help researchers to design loss more flexibly in the future.

  • (06/03/2022) One paper accepted by Pattern Recognition Journal.

  • (05/15/2022) One paper accepted by the 38th Conference on Uncertainty in Artificial Intelligence (UAI 2022).

  • (05/09/2022) One demo paper accepted by IEEE International Conference on Multimedia Information Processing and Retrieval (MIPR 2022).

  • (04/22/2022) I have passed my Ph.D. dissertation defense.

  • (04/15/2022) One paper accepted by Autonomous Driving Workshop at CVPR 2022.

  • (04/03/2022) One paper accepted by Journal of Machine Learning Research (JMLR).

  • (03/06/2022) One paper accepted by IEEE International Conference on Multimedia and Expo (ICME 2022).

  • (02/25/2022) One paper accepted by IEEE Access Journal.

  • (01/21/2022) One paper accepted by ICASSP 2022 [PDF]

  • (01/13/2022) One paper accepted by Computer Networks Journal [PDF]

  • (12/10/2021) I received the Honorable Mention of Agrusa CSE Student Innovation Competition 2021, University at Buffalo, SUNY. [Award]

  • (12/01/2021) One paper accepted by AAAI Conference on Artificial Intelligence (AAAI 2022).

  • (08/19/2021) I have passed my doctoral dissertation proposal defense.

  • (07/22/2021) One paper accepted by the International Conference on Computer Vision (ICCV 2021).

  • (06/21/2021) One paper accepted by the 2021 International Joint Conference on Biometrics (IJCB 2021).

  • (05/12/2021) I will join Robert Bosch LLC (Sunnyvale, CA) as a Deep Learning Research Intern (June 2021 - Aug 2021).

  • (03/15/2021) Our work on detecting GAN generated images using corneal reflections is reported by UB News. [Link]

  • (01/29/2021) One paper accepted by ICASSP 2021.

  • (12/11/2020) I got the first place in the graduate poster competition from the CSE department, University at Buffalo, SUNY. [Award]

  • (10/13/2020) Our work ''Exposing GAN-generated Faces Using Inconsistent Corneal Specular Highlights'' has been mentioned by Financial Times and usa-vision

Preprints

  • Rank-based Decomposable Losses in Machine Learning: A Survey [PDF]
    Shu Hu, Xin Wang, Siwei Lyu.
    ArXiv, 2022, 7.
  • GAN-generated Faces Detection: A Survey and New Perspectives [PDF]
    Xin Wang, Hui Guo, Shu Hu, Ming-Ching Chang, Siwei Lyu.
    ArXiv, 2022, 2.

Selected Papers

  • Sum of Ranked Range Loss for Supervised Learning [PDF]
    Shu Hu, Yiming Ying, Xin Wang, Siwei Lyu.
    Journal of Machine Learning Research (JMLR) , 2022, 4.
  • Eyes Tell All: Irregular Pupil Shapes Reveal GAN-generated Faces [PDF] [Twitter] [Chinese News] [Youtube] [US News 1] [US News 2]
    Hui Guo, Shu Hu, Xin Wang, Ming-Ching Chang, Siwei Lyu.
    47th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2022, 1.
  • Stochastic Planner-Actor-Critic for Unsupervised Deformable Image Registration [PDF]
    Ziwei Luo, Jing Hu, Xin Wang, Shu Hu, Bin Kong, Youbing Yin, Qi Song, Xi Wu and Siwei Lyu.
    AAAI Conference on Artificial Intelligence (AAAI), 2022, 2.
  • TkML-AP: Adversarial Attacks to Top-k Multi-Label Learning [PDF]
    Shu Hu, Lipeng Ke, Xin Wang, Siwei Lyu.
    International Conference on Computer Vision (ICCV), 2021, 10.
  • Exposing GAN-generated Faces Using Inconsistent Corneal Specular Highlights [PDF] [Project page] [Code] [Poster] [Award] [BBC] [WKBW-TV] [Futurism] [CNET]
    Shu Hu, Yuezun Li, Siwei Lyu.
    46th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2021, 1.
  • Learning by Minimizing the Sum of Ranked Range [PDF] [Code] [Poster]
    Shu Hu, Yiming Ying, Xin Wang, Siwei Lyu.
    Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS), 2020, 9.
  • Uncertainty Aware Semi-Supervised Learning on Graph Data [PDF]
    Xujiang Zhao, Feng Chen, Shu Hu, Jin-Hee Cho.
    Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS)(Spotlight), 2020, 9.