Feng Liu (he/him) -- Assistant Professor at The University of Melbourne


Home


Feng Liu

Feng Liu, Ph.D.

Assistant Professor in Machine Learning,
ARC Discovery Early Career Researcher Awardee (Machine Learning),
ACS Australasian AI Emerging Research Awardee,
School of Computing and Information Systems, The University of Melbourne

Visiting Scientist @ Imperfect Information Learning Team,
RIKEN Center for Advanced Intelligence Project (RIKEN-AIP)

Visting Fellow @ DeSI Lab,
Australian Artificial Intelligence Institute, UTS

Address: Room 3317, Level 3, Melbourne Connect (Building 290),
700 Swanston Street, University of Melbourne VIC 3010 Australia.
E-mail (academic): fengliu.ml [at] gmail.com or feng.liu1 [at] unimelb.edu.au
E-mail (industry): fengliu.genai [at] gmail.com
[Google Scholar] [Github] [Group Website] [CV]


Opportunities

  • Long-time Recruitment (PhD): I am always looking for self-motivated PhD (two per semester). Please see this page for recruiting information, and check this page for the school information. Update (Aug. 27, 2025): No PhD position is available before Jan 2028.

  • Long-time Recruitment (RA): Meanwhile, I am happy to host remote research trainees (paid). You can collaborate with many excellent researchers in the frontier machine learning research areas in our group. Check this page for more information. Update (Aug. 15, 2024): We are actively recruiting RAs. Please send your CV to my email (Gmail).

  • Project-based Recruitment:

    • No active recruitment.


Biography

    I am a machine learner with research interests in hypothesis testing and trustworthy machine learning. I am currently an ARC DECRA Fellow and an Assistant Professor in Machine Learning at the School of Computing and Information Systems, The University of Melbourne, Australia. We are also running the Trustworthy Machine Learning and Reasoning (TMLR) Lab where I am one of co-directors (see this page for details). In addition, I am a Visiting Scientist at RIKEN-AIP, Japan, and a Visting Fellow at DeSI Lab, Australian Artificial Intelligence Institute, University of Technology Sydney. I was the recipient of the Australian Laureate postdoctoral fellowship. I received my Ph.D. degree in computer science at the University of Technology Sydney in 2020, advised by Dist. Prof. Jie Lu and Prof. Guangquan Zhang. I was a research intern at the RIKEN-AIP, working on the robust domain adaptation project with Prof. Masashi Sugiyama, Dr. Gang Niu and Dr. Bo Han. I visited Gatsby Computational Neuroscience Unit at UCL and worked on the hypothesis testing project with Prof. Arthur Gretton, Dr. Danica J. Sutherland and Dr. Wenkai Xu.

    I have received the ARC Discovery Early Career Researcher Award, the FEIT Excellence Award in Early Career Research at The University of Melbourne, the Outstanding Paper Award of NeurIPS (2022), the Outstanding Reviewer Award of NeurIPS (2021), the Outstanding Reviewer Award of ICLR (2021), the UTS Best Thesis Award (Dean's list), the UTS-FEIT HDR Research Excellence Award (2019) and the Best Student Paper Award of FUZZ-IEEE (2019). My publications are mainly distributed in high-quality journals or conferences, such as Nature Plants, Nature Communications, JMLR, IEEE-TPAMI, IEEE-TNNLS, IEEE-TFS, NeurIPS, ICML, ICLR, KDD, IJCAI, and AAAI. I have served as area chairs (AC) for ICML, NeurIPS, ICLR, AISTATS, senior program committee (SPC) members for AAAI, IJCAI. I also serve as an Associate Editor for ACM TOPML, Associate Editor for IJMLC, Action Editor for Neural Networks, Action Editor for Transactions on Machine Learning Research.


Research Interests

    My general research interest lies in statistically trustworthy machine learning which aims to make current AI systems reliable with theoretical/statistical guarantees. Specifically, current research directions include statistical hypothesis testing (as a basis for theoretical/statistical guarantees), misinformation detection/defense, privacy and copyright protection in AI systems, and efficient post-training paradigm on pre-trained vision models, vision-language models, and large-language models. Detailed topics are as follows.
    Statistical Hypothesis Testing:
  • Two-sample Testing: Testing if two datasets are drawn from the same distribution.

  • (Conditional) Independence Testing: Testing if two datasets are independent (on conditions).

  • Distributional Closeness Testing: Testing if two datasets are statistically close to each other, given a pre-defined discrepancy.

    Misinformation Detection/Defense:
  • AI-generated data detection: Detecting AI-generated content.

  • Defending against Adversarial Attacks: Detecting adversarial attacks (i.e., adversarial attack detection); Training a robust model against future adversarial attacks (i.e., adversarial training); Purifying adversarial inputs.

  • Being Aware of Out-of-distribution Data: Detecting out-of-distribution data; Training a robust model in the open world (e.g., open-set learning, out-of-distribution generalization).

    Privacy and Copyright Protection in AI Systems:
  • Copyright Protection on Training Data and/or Generative Models: Detecting in-directly illegal use of data or models. Ideally, the detection methods can have a controllable false alarm rate.

  • Data Privacy Protection: Training a discriminative model to ensure that the training data will not be obtained by inverting the model (i.e., defending against model-inversion attacks).

    Efficient Post-training:
  • Parameter-efficient Post-training/Finetuning: Post-train or finetune pre-trained models with limited parameters, e.g., model reprogramming, low-rank adaptation, and prompt tuning.

  • Efficient Black-box Post-training/Finetuning: Post-train or finetune black-box models via inference-only optimization.


Research Highlights


Research Experience


Education

  • Ph.D. in Computer Science (November 2020)

  • Faculty of Engineering and Information Technology,
    University of Technology Sydney, Sydney, Australia.
    Supervised by Dist. Prof. Jie Lu and Prof. Guangquan Zhang


Sponsors

Australian Research Council CSIRO NSF