Design Methodology for Deep Out-of-Distribution Detectors in Real-Time Cyber-Physical Systems
This addresses the deployment challenge for real-time cyber-physical systems, but it is incremental as it builds on existing OOD detectors with a focus on optimization.
The paper tackles the problem of deploying out-of-distribution detectors in resource-constrained cyber-physical systems by proposing a design methodology that uses genetic algorithms and quantization to optimize for accuracy and response time, resulting in a drastic reduction in response time while maintaining comparable accuracy on embedded platforms.
When machine learning (ML) models are supplied with data outside their training distribution, they are more likely to make inaccurate predictions; in a cyber-physical system (CPS), this could lead to catastrophic system failure. To mitigate this risk, an out-of-distribution (OOD) detector can run in parallel with an ML model and flag inputs that could lead to undesirable outcomes. Although OOD detectors have been well studied in terms of accuracy, there has been less focus on deployment to resource constrained CPSs. In this study, a design methodology is proposed to tune deep OOD detectors to meet the accuracy and response time requirements of embedded applications. The methodology uses genetic algorithms to optimize the detector's preprocessing pipeline and selects a quantization method that balances robustness and response time. It also identifies several candidate task graphs under the Robot Operating System (ROS) for deployment of the selected design. The methodology is demonstrated on two variational autoencoder based OOD detectors from the literature on two embedded platforms. Insights into the trade-offs that occur during the design process are provided, and it is shown that this design methodology can lead to a drastic reduction in response time in relation to an unoptimized OOD detector while maintaining comparable accuracy.