CauseRL: Reinforcement Learning-based Random Walks for Root Cause Analysis in Microservices

Key Points

• This research aims to develop a reinforcement learning framework for identifying root causes in microservices.
• Implemented a reinforcement learning-based algorithm for root cause analysis.
• Utilized random walks to explore potential causes.
• Tested the framework in simulated microservices environments.
• Showed improved accuracy in identifying root causes compared to traditional methods.
• Reduced time taken for troubleshooting by streamlining the analysis process.