We present a deep learning framework for probabilistic pixel-wise semantic segmentation, which we term Bayesian SegNet. Semantic segmentation is an important tool for visual scene understanding and a meaningful measure of uncertainty is essential for decision making. Our contribution is a practical system which is able to predict pixelwise class labels with a measure of model uncertainty. We ac...

This paper reviews research in relation with modelling uncertainty within Decision Support Systems (DSS) from 2000 to 2011. It specifically addresses software that has been built or prototyped with the purpose of supporting actual decision making, which is able to explicitly deal with uncertainty (widely understood) on the corresponding model parameters and/or data. The main DSS features analys...

Mining companies world-wide are faced with the problem of how to accurately value and plan extraction projects subject to uncertainty in both future price and ore grade. Whilst the methodology of modelling price uncertainty is reasonably well understood, modelling ore-grade uncertainty is a much harder problem to formulate, and when attempts have been made the solutions take unfeasibly long tim...

Abstract Uncertainty can be classified as either aleatoric (intrinsic randomness) or epistemic (imperfect knowledge of parameters). The majority frameworks assessing infectious disease risk consider only uncertainty. We ever observe a single epidemic, and therefore cannot empirically determine Here, we characterise both uncertainty using time-varying general branching process. Our framework exp...

Abstract A novel solution to the smoothing problem for multi-object dynamical systems is proposed and evaluated. The of interest contain an unknown varying number objects that are partially observed under noisy corrupted observations. In order account lack information about different aspects this type complex system, alternative representation uncertainty based on possibility theory considered....