This paper proposes a feature extraction and fusion methodology to perform fault detection and classification in distributed physical processes generating heterogeneous data. The underlying concept is built upon a semantic framework for multi-sensor data interpretation using graphical models of Probabilistic Finite State Automata (PFSA). While the computational complexity is reduced by pruning the fused graphical model using an information-theoretic approach, the algorithms are developed to achieve high reliability via retaining the essential spatiotemporal characteristics of the physical processes. The concept has been validated on a simulation test bed of distributed shipboard auxiliary systems.

No abstract

With the advent of Deep Learning, the field of machine learning (ML) has surpassed human-level performance on diverse classification tasks. At the same time, there is a stark need to characterize and quantify reliability of a model's prediction on individual samples. This is especially true in applications of such models in safety-critical domains of industrial control and healthcare. To address this need, we link the question of reliability of a model's individual prediction to the epistemic uncertainty of the model's prediction. More specifically, we extend the theory of Justified True Belief (JTB) in epistemology, created to study the validity and limits of human-acquired knowledge, towards characterizing the validity and limits of knowledge in supervised classifiers. We present an analysis of neural network classifiers linking the reliability of its prediction on a test input to characteristics of the support gathered from the input and hidden layers of the network. We hypothesize that the JTB analysis exposes the epistemic uncertainty (or ignorance) of a model with respect to its inference, thereby allowing for the inference to be only as strong as the justification permits. We explore various forms of support (for e.g., k-nearest neighbors (k-NN) and ℓp-norm based) generated for an input, using the training data to construct a justification for the prediction with that input. Through experiments conducted on simulated and real datasets, we demonstrate that our approach can provide reliability for individual predictions and characterize regions where such reliability cannot be ascertained.

No abstract

No abstract