Abstract
This paper evaluates the efficacy of a machine learning approach to data fusion using convolved multi-output Gaussian processes in the context of geological resource modeling. It empirically demonstrates that information integration across multiple information sources leads to superior estimates of all the quantities being modeled, compared to modeling them individually. Convolved multi-output Gaussian processes provide a powerful approach for simultaneous modeling of multiple quantities of interest while taking correlations between these quantities into consideration. Experiments are performed on large scale data taken from a mining context.
Highlights
Gaussian processes (GPs) [Rasmussen and Williams, 2006] are powerful non-parametric Bayesian learning techniques that can handle correlated, uncertain and incomplete data
This paper evaluates a machine learning approach to data fusion using a convolved GP model applied to a multioutput problem of geological resource modeling; multiple kernel combinations including the best results from past works mentioned above are compared
This paper empirically studied the problem of geological resource modeling using a machine learning approach to convolved multi-output Gaussian processes (MOGPs)
Summary
Gaussian processes (GPs) [Rasmussen and Williams, 2006] are powerful non-parametric Bayesian learning techniques that can handle correlated, uncertain and incomplete data. GPs yield a continuous domain representation of the data and can be sampled at any desired resolution (multi-scale model). They model and use the spatial correlation of the given data to estimate the values for unknown points of interest. Data fusion in the context of Gaussian processes is necessitated by the presence of multiple, multi-sensor, multi-attribute, incomplete and/or uncertain data sets of the entity being modeled. This paper presents an empirical evaluation of a machine learning approach to performing data fusion with Gaussian processes (based on convolved GPs) in the context of vector valued data. Experiments are performed on large scale data obtained from a mining context
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have