Monitor and Recover: A Paradigm for Future Research on Distribution Shift in Learning-Enabled Cyber-Physical Systems
This work tackles reliability issues in cyber-physical systems under distribution shift, presenting a novel paradigm shift rather than incremental improvements.
The paper addresses the vulnerability of neural networks to distribution shift in learning-enabled cyber-physical systems by proposing a monitor and recover paradigm, which focuses on robust safety monitoring and recovery instead of detection and abstention.
With the known vulnerability of neural networks to distribution shift, maintaining reliability in learning-enabled cyber-physical systems poses a salient challenge. In response, many existing methods adopt a detect and abstain methodology, aiming to detect distribution shift at inference time so that the learning-enabled component can abstain from decision-making. This approach, however, has limited use in real-world applications. We instead propose a monitor and recover paradigm as a promising direction for future research. This philosophy emphasizes 1) robust safety monitoring instead of distribution shift detection and 2) distribution shift recovery instead of abstention. We discuss two examples from our recent work.