Trusted Confidence Bounds for Learning Enabled Cyber-Physical Systems
This work addresses the problem of ensuring safety and transparency in cyber-physical systems for engineers and developers, though it is incremental as it builds on existing conformal prediction methods.
The paper tackles the challenge of integrating learning-enabled components into safety-critical cyber-physical systems by developing an approach to compute trusted confidence bounds for neural network predictions, using Inductive Conformal Prediction and a Triplet Network to estimate similarity and achieve efficient real-time performance on a robotic navigation benchmark.
Cyber-physical systems (CPS) can benefit by the use of learning enabled components (LECs) such as deep neural networks (DNNs) for perception and decision making tasks. However, DNNs are typically non-transparent making reasoning about their predictions very difficult, and hence their application to safety-critical systems is very challenging. LECs could be integrated easier into CPS if their predictions could be complemented with a confidence measure that quantifies how much we trust their output. The paper presents an approach for computing confidence bounds based on Inductive Conformal Prediction (ICP). We train a Triplet Network architecture to learn representations of the input data that can be used to estimate the similarity between test examples and examples in the training data set. Then, these representations are used to estimate the confidence of set predictions from a classifier that is based on the neural network architecture used in the triplet. The approach is evaluated using a robotic navigation benchmark and the results show that we can computed trusted confidence bounds efficiently in real-time.