A sequential decision problem, based on the task of identifying the species of trees given acoustic echo data collected from them, is considered with well-known stochastic classifiers, including single and mixture Gaussian models. Echoes are processed with a preprocessing stage based on a model of mammalian cochlear filtering, using a new discrete low-pass filter characteristic. Stopping time p...

In this paper, we explore a sequential Bayesian bound for state-space models focusing on hybrid continuous and discrete random states. We provide an analytic recursion for the sequential Weiss–Weinstein (SWW) bound for linear state-space models with solutions for Gaussian, uniform, and exponential distributions as derived, as well as for a combination of these. We compare the SWW bound for disc...

We consider the problem of sequential learning from categorical observations bounded in [0, 1]. We establish an ordering between the Dirichlet posterior over categorical outcomes and a Gaussian posterior under observations with N(0, 1) noise. We establish that, conditioned upon identical data with at least two observations, the posterior mean of the categorical distribution will always second-o...

In mining projects, all uncertainties associated with a project must be considered to determine the feasibility study. Grade uncertainty is one of the major components of technical uncertainty that affects the variability of the project. Geostatistical simulation, as a reliable approach, is the most widely used method to quantify risk analysis to overcome the drawbacks of the estimation methods...

A deep Gaussian process (DGP) is a deep network in which each layer is modelled with a Gaussian process (GP). It is a flexible model that can capture highly-nonlinear functions for complex data sets. However, the network structure of DGP often makes inference computationally expensive. In this paper, we propose an efficient sequential inference framework for DGP, where the data is processed seq...