Shuaicheng Niu

 
College of Computing and Data Science, Nanyang Technological University
Address: ABN-03A-04, 61 Nanyang Drive Singapore 637335
Email : niushuaicheng [at] gmail.com     Github     Google Scholar

About Me

I am a research fellow at College of Computing and Data Science, Nanyang Technological University, working with Prof. Miao Chun Yan. I received my PhD degree from the South China University of Technology under the supervision of Prof. Mingkui Tan, where I also collaborated closely with Dr. Jiaxiang Wu and Dr. Peilin Zhao from Tencent AI Lab. My research interests are broadly in machine learning and mainly focus on:


  • Efficient pre-trained model re-use
  • Test-time/inference-time learning/computing/adaptation/editing for discriminative or generative models
  • Out-of-Distribution Generalization: stable, on-device, forward-only, lightweight, black-box, edge-edge collaborative and quantized model adaptations in the open wild wrold
  • Automated Machine Learning: neural architecture search, discrete optimization
  • Deep Learning Applications on various tasks

News

  • 2025.02: One paper accepted by CVPR 2025!
  • 2025.01: One paper accepted by ICLR 2025!
  • 2024.12: One paper accepted by AAAI 2025!
  • 2024.09: One paper accepted by NeurIPS 2024!
  • 2024.07: One paper accepted by ECCV 2024!
  • 2024.05: One paper accepted by ICML 2024 as Oral (Top 1.5%)!
  • 2024.01: One paper accepted by ICLR 2024!
  • 2023.09: One paper accepted by NeurIPS 2023!
  • 2023.07: One paper accepted by ACM MM 2023!
  • 2023.06: Successfully accomplished my PhD thesis defense!
  • 2023.05: One paper accepted by IEEE TMM!
  • 2023.03: Last day at Tencent. Thanks to everything here!
  • 2023.01: One paper is accepted by ICLR 2023 as Oral (Top 5% among accepted papers)! Happy Chinese New Year!
  • 2022.12: We are the second place in AutoML Decathlon 2022 (NeurIPS Competition Track). Thanks to my TEG-AutoML teammates!
  • 2022.07: One collaborative paper is accepted by ECCV 2022!
  • 2022.05: One paper is accepted by ICML 2022!
  • 2022.04: Two papers w.r.t. boosting test-time performance and test-time adaptation are released to arXiv!
  • 2022.03: I have been invited as a reviewer for NeurIPS 2022, ECCV 2022 and TNNLS.
  • 2022.01: I have been invited as a reviewer for ICML 2022.
  • 2021.12: I have been invited as a reviewer for CVPR 2022.
  • 2021.10: I won the President Scholarship of SCUT!
  • 2021.11: One paper is accepted by Neural Networks 2021.
  • 2021.07: Our paper is accepted by ACM MM 2021 as Oral!
  • 2021.05: Our paper is accepted by ICML 2021 as Spotlight!
  • 2021.05: Our paper is accepted by IJCAI 2021.
  • 2021.03: I am a Ph.D candidate now!
  • 2021.03: Our paper is accepted by CVPR 2021!
  • 2020.12: I have been invited as a reviewer for TIP.
  • 2020.07: One paper is accepted by TIP 2020!
  • 2020.04: One paper for tackling COVID-19 is released to arXiv. Hope everyone well under the epidemic!
  • 2019.11: Our paper is accepted by TKDE 2019.
  • 2019.08: Start my internship at Tencent AI Lab in Shenzhen.
  • 2018.09: Begin my journey in SMIL Lab, South China University of Technology at Guangzhou.

Selected Papers [Full List]


-----------------------Conference Papers-----------------------

Test-Time Model Adaptation with Only Forward Passes
Shuaicheng Niu, Chunyan Miao, Guohao Chen, Pengcheng Wu, Peilin Zhao
International Conference on Machine Learning (ICML), 2024. Oral (Top 1.5%).
[arXiv] [Code]
Towards Stable Test-Time Adaptation in Dynamic Wild World
Shuaicheng Niu, Jiaxiang Wu, Yifan Zhang, Zhiquan Wen, Yaofo Chen, Peilin Zhao, Mingkui Tan
International Conference on Learning Representations (ICLR), 2023. Oral, top-5% among accepted papers
[Paper] [Code] [机器之心/极市平台/将门创投/etc.]
Efficient Test-Time Model Adaptation without Forgetting
Shuaicheng Niu, Jiaxiang Wu, Yifan Zhang, Yaofo Chen, Shijian Zheng, Peilin Zhao, Mingkui Tan
International Conference on Machine Learning (ICML), 2022.
[Paper] [Code] [Poster]
Cross-Device Collaborative Test-Time Adaptation
Guohao Chen*, Shuaicheng Niu (co-first)*, Deyu Chen, Shuhai Zhang, Changsheng Li, Yuanqing Li, Mingkui Tan
Advances on Neural Information Processing Systems (NeurIPS), 2024.
[Paper] [Code]
Learning to Generate Gradients for Test-Time Adaptation
Qi Deng*, Shuaicheng Niu (co-first)*, Ronghao Zhang, Yaofo Chen, Runhao Zeng, Jian Chen, Xiping Hu
Annual AAAI Conference on Artificial Intelligence (AAAI), 2025.
[Paper]
AdaXpert: Adapting Neural Architecture for Growing Data
Shuaicheng Niu, Jiaxiang Wu, Guanghui Xu, Yifan Zhang, Yong Guo, Peilin Zhao, Peng Wang, Mingkui Tan
International Conference on Machine Learning (ICML), 2021.
[Paper] [Code]
Towards Accurate Text-based Image Captioning with Content Diversity Exploration
Guanghui Xu*, Shuaicheng Niu (co-first)*, Mingkui Tan, Yucheng Luo, Qing Du, Qi Wu
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021.
[Paper] [Code]
Source-free Domain Adaptation via Avatar Prototype Generation and Adaptation
Zhen Qiu, Yifan Zhang, Hongbin Lin, Shuaicheng Niu, Yanxia Liu, Qing Du, Mingkui Tan
International Joint Conference on Artificial Intelligence (IJCAI), 2021.
[Paper] [Code]

-------------------------Journal Papers------------------------

Disturbance-immune weight sharing for neural architecture search
Shuaicheng Niu, Jiaxiang Wu, Yifan Zhang, Yong Guo, Peilin Zhao, Junzhou Huang, Mingkui Tan
Neural Networks (NN), 2021.
[Paper]
Collaborative Unsupervised Domain Adaptation for Medical Image Diagnosis
Yifan Zhang, Ying Wei, Qingyao Wu, Peilin Zhao, Shuaicheng Niu, Junzhou Huang, Mingkui Tan
IEEE Transactions on Image Processing (TIP), 2020.
[Paper] [Code]
Online Adaptive Asymmetric Active Learning with Limited Budgets
Yifan Zhang*, Peilin Zhao*, Shuaicheng Niu (co-first)*, Qingyao Wu, Jiezhang Cao, Junzhou Huang, Mingkui Tan
IEEE Transactions on Knowledge and Data Engineering (TKDE), 2019.
[Paper] [Code]

--------------------------Pre-print Papers------------------------

Boost Test-Time Performance with Closed-Loop Inference
Shuaicheng Niu, Jiaxiang Wu, Yifan Zhang, Guanghui Xu, Haokun Li, Junzhou Huang, Yaowei Wang, Mingkui Tan
arXiv:2203.10853, 2022.
[arXiv]
COVID-DA: Deep Domain Adaptation from Typical Pneumonia to COVID-19
Yifan Zhang, Shuaicheng Niu, Zhen Qiu, Ying Wei, Peilin Zhao, Jianhua Yao, Junzhou Huang, Qingyao Wu, Mingkui Tan
arXiv:2005.01577, 2020.
[arXiv] [Code]


Professonal Activities

  • IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), Reviewer
  • IEEE Transactions on Medical Imaging (TMI), Reviewer
  • IEEE Transactions on Image Processing (TIP), Reviewer
  • IEEE Transactions on Neural Networks and Learning Systems (TNNLS), Reviewer
  • International Journal of Computer Vision (IJCV), Reviewer
  • International Conference on Machine Learning (ICML), Reviewer 22, 23, 24, 25
  • International Conference on Learning Representations (ICLR), Reviewer 23, 24, 25
  • Advances on Neural Information Processing Systems (NeurIPS), Reviewer 22, 23, 24
  • The Conference on Computer Vision and Pattern Recognition (CVPR), Reviewer 22, 23, 24, 25
  • European Conference on Computer Vision (ECCV), Reviewer 22, 24
  • International Conference on Computer Vision (ICCV), Reviewer 23

Experiences

    SMIL Lab, South China University of Technology, Guangzhou, China

    Research Assistant,   Sep. 2018 ~ June 2023

    Supervisor: Prof. Mingkui Tan

    Tencent AI Lab, Shenzhen, China

    Research Intern,   August 2019 ~ March 2023

    Mentors: Dr. Jiaxiang Wu and Dr. Peilin Zhao

------------------------------------

^_^ The source code is adapted from both Jon Barron and Can Qin.