Multi-agent reinforcement learning-based passenger spoofing attack on mobility-as-a-service

HIGHLIGHTS

• What: The authors investigate the threat in the smart coordinator and identify a new reinforcement learningbased attack, named passenger spoofing attack (PSA), which leverages the diversity of passengers and the heterogeneous service planning to prioritize malicious agents in the journey planning process. The study reveals conditions where transportation modes are excluded, providing insights into system behavior and cost implications. While there have been studies on reinforcement learning-based attacks, such as the work discussed, these primarily focus on data poisoning attacks in crowdsensing systems rather than addressing the specific threat to passenger heterogeneity in Journey Planning Problem . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or Email

  Password

  Remember me

  Lost your password?

  If you don't have an account, you can create one here.