Yanxiao Liu(刘延骁)
 |
Yanxiao Liu |
About me
Before I join The Chinese University of Hong Kong, I received my B.Eng. degree in Electronic Information Engineering and minor degree in Philosophy from The Chinese University of Hong Kong (Shenzhen) in 2021. From January 2024 to June 2024, I visited Stanford University, ISL Lab, hosted by Professor Ayfer Özgür and supported by PhD IMPAC Award 2023–24. I will graduate in summer 2025 and I am looking for postdoc positions.
Research
Advisors: Professor Cheuk Ting Li and Professor Raymond W. Yeung
My research interests lie in network information theory and applications. With more details, I currently work on the following directons:
One-shot/finite-blocklength network information theory:
I prove one-shot achievability results, second-order results and asymptotic inner/outer bounds on classical network information theory problems. Currently, my primary focus is on lossy compression and interference channels.
Machine learning and privacy:
I am interested in the theoretical understanding of machine learning problems through the application of information theory techniques. I am currently employing channel simulation tools to distributed learning and differential privacy.
Modern wireless communications:
I deveolop fundamental theory and low-complexity algorithms for large-scale networks in challenging environments. For instance, I design practical algorithms that jointly solve the multiflow, scheduling and network coding problems in large-scale networks, offering provable performance guarantees.
Awards
NeurIPS Scholar Award (2024)
PhD International Mobility for Partnerships and Collaborations Award (PhD IMPAC Award 2023–24), highest supporting rate.
CUHK Postgraduate Student Scholarship 2021-2025
News
Nov 2024: One paper has been submitted to Transactions on Information Theory.
Sep 2024: One paper has been accepted by NeurIPS 2024 ([poster] and [Slides])
Jul 2024: One paper has been submitted to INFOCOM 2025 (under double-blind peer-review).
May 2024: One paper has been accepted by Transactions on Information Theory.