While many deep learning (DL)-based networking systems have demonstrated superior performance, the underlying Deep Neural Networks (DNNs) remain blackboxes and stay uninterpretable for network operators. The lack of interpretability makes DL-based networking systems prohibitive to deploy in practice. In this paper, we propose Metis, a framework that provides interpretability for two general categories of networking problems spanning local and global control. Accordingly, Metis introduces two different interpretation methods based on decision tree and hypergraph, where it converts DNN policies to interpretable rule-based controllers and highlight critical components based on analysis over hypergraph. We evaluate Metis over two categories of state-of-the-art DL-based networking systems and show that Metis provides human-readable interpretations while preserving nearly no degradation in performance. We further present four concrete use cases of Metis, showcasing how Metis helps network operators to design, debug, deploy, and ad-hoc adjust DL-based networking systems.


Interpreting Deep Learning-Based Networking Systems

Zili Meng, Minhu Wang, Jiasong Bai, Mingwei Xu, Hongzi Mao, Hongxin Hu
Proceedings of the 2020 ACM SIGCOMM Conference


  title={Interpreting Deep Learning-Based Networking Systems},
  author={Meng, Zili and Wang, Minhu and Bai, Jiasong and Xu, Mingwei and Mao, Hongzi and Hu, Hongxin},
  booktitle={Proc. ACM SIGCOMM},


SIGCOMM video (20min version, with subtitles)
SIGCOMM video (10min version, with subtitles)
Talk at the APNet Workshop 2020
Topic preview given by Keith Winstein


Presentation Slides (.pptx, better played with Windows and Office 2019+)
Presentation Slides (.pptx, compatible version for Mac users)





The research is supported by the National Natural Science Foundation of China (No. 61625203 and 61832013), and the National Key R&D Program of China (No. 2017YFB0801701).


For any questions, please send an email to zilim@ieee.org.