Research Assistant, Machine Learning Safety
NTU Main Campus, Singapore
School of Computer Science and Engineering invites applications for the position of Research Assistant.
In this research project, the research team will explore the problem of achieving safety assurance for learning enabled components in safety-critical cyber-physical systems. A classic example is the object detection and classification component in an autonomous vehicle.
In this project, the research team specifically focus on two research problems in this area:
- Detecting out-of-distribution images efficiently at runtime so that appropriate fail-safe operations can be performed when images outside of the training space are encountered
- Probabilistic formal verification of learning enabled components using scenario optimization so that a high performance and high assurance safety controller can be designed
- Acquire knowledge and understand the key technologies developed for safety assurance in ML, particularly related to the problem of out-of-distribution (OOD) detection
- Implement technologies developed in the research project on an autonomous vehicle (AV) simulation platform (Carla)
- Implement technologies developed in the research project on an embedded test-bed (Duckietown)
- Contribute towards preparation of research manuscripts
- Work with a globally renowned research team in cyber-physical systems (https://cps-research-group.github.io/)
- Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, Electronics Engineering, Instrumentation Engineering or related fields
- Proficient in programming with C and Python
- Prior design and implementation experience in embedded systems’ technologies
- Prior design and implementation experience in machine learning technologies, particularly for vision related applications
- Prior research experience in safety assurance, embedded systems or machine learning is a plus
We regret that only shortlisted candidates will be notified.