I am a networking researcher at Microsoft Research Redmond and the Office of the CTO, Azure for Operators. My research interests span:

  • practical reinforcement learning (RL) for networking
  • video systems (streaming, conferencing, etc.)
  • transport and application layers

I completed my Ph.D. in computer science at Stanford University, advised by Keith Winstein and Philip Levis. My dissertation was on the development of platforms and algorithms to achieve practical RL on the internet, in the context of video streaming and congestion control. My work has received the IRTF Applied Networking Research Prize, USENIX NSDI Community Award, and USENIX ATC Best Paper Award. I am the creator of Puffer, a live-streaming site that has reached more than 200,000 users.

Before that, I graduated from Tsinghua University, where I received a B.S. in computer science from Yao Class and a B.A. in economics. I also studied at MIT in 2014.

I tweet research internship opportunities and bad jokes.

Nov 2022
Serving on the PC of CoNEXT 2023.
Oct 2022
Serving on the award committee of the IRTF Applied Networking Research Prize 2023. Nominations are accepted until Nov 18, 2022. How to nominate?
Sept 2022
We are hiring research interns to work with us in person next summer! We are also hiring full-time researchers. Links to apply can be found here.
May 2022
Genet is accepted into SIGCOMM 2022! It uses curriculum learning to improve the performance and generalization of RL-based network algorithms, taking another (small) step toward practical RL in networking.
Apr 2022
Serving on the PC of NSDI 2023.
Mar 2022
Serving as the Networking Area Co-Chair of the Journal of Systems Research. Check out the Call for Papers and send us your best work!
Recent Research
Zhengxu Xia*, Yajie Zhou*, Francis Y. Yan, Junchen Jiang (*equal contribution)
ACM SIGCOMM, August 2022
Genet is a novel training framework that enhances the performance and generalization of reinforcement learning (RL) algorithms in networking. Genet builds on curriculum learning with judicious use of rule-based baselines. It substantially improves the performance and generalization of simulation-trained RL algorithms under unseen workloads and in real environments.
Jeongyoon Eo, Zhixiong Niu, Wenxue Cheng, Francis Y. Yan, Rui Gao, Jorina Kardhashi, Scott Inglis, Michael Revow, Byung-Gon Chun, Peng Cheng, Yongqiang Xiong
Asia-Pacific Workshop on Networking (APNet), July 2022
emoji_events Best Paper Award
OpenNetLab is an open platform for training, validating, and evaluating RL-based congestion-control (bandwidth estimation) algorithms for real-time communications (RTC) such as videoconferencing. It has successfully aided the development of novel RL-based congestion-control algorithms for RTC during our Grand Challenge hosted at ACM MMSys '21.
Zhiying Xu, Francis Y. Yan, Rachee Singh, Justin T. Chiu, Alexander M. Rush, Minlan Yu
arXiv:2210.13763 (preprint), October 2022
Teal is a deep learning-based traffic engineering (TE) scheme that accelerated the TE optimization on large WANs by several orders of magnitude while achieving near-optimal traffic allocation.
Michael Rudow, Francis Y. Yan, Abhishek Kumar, Ganesh Ananthanarayanan, Martin Ellis, K.V. Rashmi
In submission
Tambur is a new approach to forward error correction (FEC) for videoconferencing built upon streaming codes and machine learning. Tambur reduced decoding failures while consuming less bandwidth for redundancy.
Francis Y. Yan, Hudson Ayers, Chenzhi Zhu, Sadjad Fouladi, James Hong, Keyi Zhang, Philip Levis, Keith Winstein
USENIX Symposium on Networked Systems Design and Implementation (NSDI), February 2020
We built Puffer, a free, publicly accessible website that live-streams television channels and operates as a randomized experiment of adaptive bitrate (ABR) algorithms. As of June 2020, Puffer has attracted 120,000 real users and streamed 60 years of video across the internet. Using Puffer, we developed an ML-based ABR algorithm, Fugu, that robustly outperformed existing schemes by learning in situ, on real data from its actual deployment environment.
Francis Y. Yan, Jestin Ma, Greg D. Hill, Deepti Raghavan, Riad S. Wahby, Philip Levis, Keith Winstein
USENIX Annual Technical Conference (ATC), July 2018
Pantheon is a “training ground” for congestion-control research and has assisted four schemes from other research groups in publishing at NSDI 2018 (Copa and Vivace), ICML 2019 (Aurora), and SIGCOMM 2020 (TCP-TACK). It also enabled our own ML-based congestion-control algorithm, Indigo, which was trained to imitate expert congestion-control algorithms we created in emulation and achieved good performance over the real internet.
Professional Activities
  • Program Committee Member, ACM CoNEXT 2023
  • Award Committee Member, IRTF Applied Networking Research Prize 2023
  • Program Committee Member, USENIX NSDI 2023
  • Networking Area Chair, Journal of Systems Research (2022–2023)
  • Award Committee Member, IRTF Applied Networking Research Prize 2022
  • External Reviewer, USENIX NSDI 2022
  • Editorial Board Member, Journal of Systems Research (2021–2022)
  • Session Chair and Challenge Chair, ACM MMSys 2021
  • Organizer, Grand Challenge on Bandwidth Estimation for Real-Time Communications, ACM MMSys 2021
  • Reviewer, IEEE/ACM Transactions on Networking (2019–)
  • Reviewer, ACM SIGCOMM Computer Communication Review (2019)