The 21st International Conference on Formal Engineering Methods

ICFEM 2019

November 5th-9th, 2019, Shenzhen, China

International Workshop on Artificial Intelligence and Formal Methods (AI&FM 2019)

Nov. 5th, 2019

Background and Objectives

Artificial Intelligence (AI) has achieved significant progress in the past few years, and is being deployed broadly in many areas such as robotics, healthcare, self-driving, finance, etc. Formal method (FM) is a collection of techniques for the specification, development, and verification of systems in mathematically rigorous ways. FM has been successfully applied to provide assurance to the safety and correctness of critical systems, spanning from avionic systems, insulin pump, to microchips.
Up to now, the development of learning-enabled AI systems is often referred as “magic” due to its lack of mathematical rigor. In a more general sense, both quantitative and qualitative tools are in great demand to understand the problems like robustness, security, ethics and privacy raised by AI systems, and to assess the benefits of deploying AI to the whole society. On the other hand, the FM methods often suffer from state-explosion problem and thus high-performance implementations of the methods are needed. This workshop aims to nurture a community with experts of both AI and FM, from both the academia and the industry, to discuss important questions such as how to develop novel FM methods to support the development, use and governance of AI, how to apply AI to improve the performance of FM methods, etc.

Invited Talks

  • Dongdong An - A Modeling Framework of Intelligent Transportation Systems with Driver Behavior Classification based on Machine Learning (detail)
  • Jeff Cao - AI Ethics and Ethics by Design (detail)
  • Weipeng Cao - Inductive Bias and Meta Learning: A New Perspective of Formal Methods Recommendation (detail)
  • Zheng Hu - The Key Challenges of RAMS: When Huawei Full Stack and All Scenarios AI comes (detail)
  • Jay Hoon Jung and YoungMin Kwon - Defining Robustness in the Input Space Using Linear Regions (detail)
  • Wang Lin - Robustness Verification of Classification Deep Neural Networks via Linear Programming (detail)
  • Meng Sun - Using RNN Predict Tactics for Theorem Proving
  • Bai Xue - Safe Inputs Approximation for Black-Box Systems (detail)
  • Pengfei Yang - Analyzing Deep Neural Networks with Symbolic Propagation: Towards Higher Precision and Faster Verification (detail)

Program

Here is a tentative program.

Conference Organizing Committee

Program Co-Chairs

  • Xiaowei Huang, University of Liverpool, UK
  • Jiaxiang Liu, Shenzhen University, China
  • Brian Tse, Partnership on AI, USA
  • Min Zhang, East China Normal University, China

Program Committee

  • Lei Bu, Nanjing University, China
  • Liqian Chen, National University of Defense Technology, China
  • Yunwei Dong, Northwestern Polytechnical University, China
  • Xiaowei Huang, University of Liverpool, UK
  • Qin Li, East China Normal University, China
  • Jiaxiang Liu, Shenzhen University, China
  • Wenjie Ruan, Lancaster University, UK
  • Meng Sun, Peking University, China
  • Youcheng Sun, Queen's University Belfast, UK
  • Brian Tse, Partnership on AI, USA
  • Jingyi Wang, National University of Singapore, Singapore
  • Min Wu, University of Oxford, UK
  • Zhilin Wu, Chinese Academy of Sciences, China
  • Yingfei Xiong, Peking University, China
  • Zhiwu Xu, Shenzhen University, China
  • Pengfei Yang, Chinese Academy of Sciences, China
  • Zijiang Yang, Western Michigan University, USA
  • Lijun Zhang, Chinese Academy of Sciences, China
  • Miaomiao Zhang, Tongji University, China
  • Min Zhang, East China Normal University, China