Visible to the public Trusted Confidence Bounds for Learning Enabled Cyber-Physical Systems

TitleTrusted Confidence Bounds for Learning Enabled Cyber-Physical Systems
Publication TypeConference Paper
Year of Publication2020
AuthorsBoursinos, Dimitrios, Koutsoukos, Xenofon
Conference Name2020 IEEE Security and Privacy Workshops (SPW)
Date PublishedMay 2020
ISBN Number978-1-7281-9346-5
Keywordsassurance monitoring, conformal prediction, cps privacy, cyber physical systems, Cyber-physical systems, decision making, deep neural networks, Human Behavior, human factors, Monitoring, Navigation, Neural networks, privacy, pubcrawl, Real-time Systems, robot navigation, Task Analysis, triplet

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.

Citation Keyboursinos_trusted_2020