This paper examines parametric estimating techniques applied to technology-driven projects in various industry sectors. Parametric estimating is the use of a subset of independent variables to predict the cost of a project. These independent variables may be specifications, features, functions, or some other high level descriptive elements that are used to define the scope of the deliverables a...

Learning complex control policies from non-linear and redundant sensory input is an important challenge for reinforcement learning algorithms. Non-parametric methods that approximate values functions or transition models can address this problem, by adapting to the complexity of the data set. Yet, many current non-parametric approaches rely on unstable greedy maximization of approximate value f...

In this paper, a history-based parametric method was proposed and implemented. However, the size of each translator that exchanges the feature and parametric information tends to be large, due to the implementation of duplicated functions. Furthermore, because a history-based parametric translator uses a procedural model as a neutral format, it requires a geometric modelling kernel to generate ...

Nonsmooth equation-solving and optimization algorithms which require local sensitivity information are extended to systems with nonsmooth parametric differential-algebraic equations embedded. Nonsmooth differential-algebraic equations refers here to semi-explicit differential-algebraic equations with algebraic equations satisfying local Lipschitz continuity and differential right-hand side func...

Software, databases, and methods from previous predictive model studies were used to assess benefits of augmenting parametric methods with kinetic model trajectory predictions. Those results were consistent with other studies of models for aircraft position prediction accuracy during climbs and descents. A kinetic model reduced error for average arrival meter fix crossing time predictions acros...

In this article we introduce the functions Fi(x1, x2)λ1,..., λs and Th(x1, x2, x3)iλ1,..., λ2s (i = 1, 2, 3), of the word variables xi and of the natural number variables λi, where s ≥ 0. By means of these functions, we give exactly the general solution (i.e. the set of all the solutions) of the first basic parametric equation: x1x2x3x4 = x3x1λx2x5, in a free monoid.

Finding good approximations to state-action value functions is a central problem in model-free on-line reinforcement learning. The use of non-parametric function approximators enables us to simultaneously represent model and confidence. Since Q functions are usually discontinuous, we present a novel Gaussian process (GP) kernel function to cope with discontinuity. We use a manifold-based distan...

This is an extension of earlier work on vocal tract area function modelling (Fant, 1960, 1992, 1993; Lin, 1990) retaining a minimum of three independent control parameters in a more complex, physiologically oriented model providing a flexible choice of detail structure. Systematic perturbation analysis of vocal tract boundary conditions and of values of the independent control parameters have b...