We examine Bayesian methods for learn ing Bayesian networks from a combination of prior knowledge and statistical data In particular we unify the approaches we pre sented at last year s conference for discrete and Gaussian domains We derive a gen eral Bayesian scoring metric appropriate for both domains We then use this metric in combination with well known statistical facts about the Dirichlet...

Background and purpose — Machine learning (ML) techniques are a form of artificial intelligence able to analyze big data. Analyzing the outcome (digital) questionnaires, ML might recognize different patterns in answers that relate types pathology. With this study, we investigated proof-of-principle ML-based diagnosis patients with hip complaints using digital questionnaire Kellgren Lawrence (KL...

The localization of plastic deformation into a shear band is discussed as an instability of -plastic flow and a precursor to rupture. Experimental observations are reviewed, a ge~eral theoretical framework is presented, and specific calculations of critical conditions are carried out for a variety of material models. The interplay between features of inelastic constitutive description~ especial...

The paper concerns the probabilistic eval­ uation of plans in the presence of unmea­ sured variables, each plan consisting of sev­ eral concurrent or sequential actions. We establish a graphical criterion for recogniz­ ing when the effects of a given plan can be predicted from passive observations on measured variables only. When the crite­ rion is satisfied, a closed-form expression is provide...

Energy minimization e orts to predict polypeptide structures assume their native conformation corre sponds to the global minimum free energy state Given this assumption the problem becomes that of develop ing e cient global optimization techniques applicable to polypeptide energy models This general structure prediction objective is also known as the protein fold ing problem Our prediction algo...

Many applications require that we learn the pa­ rameters of a model from data. EM (E xpectation­ Maximization) is a method for learning the pa­ rameters of probabilistic models with missing or hidden data. There are instances in which this method is slow to converge. Therefore, sev­ eral accelerations have been proposed to improve the method. None of the proposed acceleration methods are theore...

Modeling support for dynamic simulation of chemical-process flowsheets, which is of significant ®alue for plantwide dynamic simulation using differential ] algebraic model formulations, is to date ®ery limited when one or more unit models include partial differential equations. Se®eral new techniques that pro®ide modeling support for such simulations are presented. These techniques are based on...

We hy poth e sized that scores on the Im plicit As so ci a tion Test (IAT) are con founded with a gen eral cog ni tive abil ity of how quickly one can pro cess in for ma tion when the IAT cat e go ries seem in con gru ent com pared to when they are con gru ent. Across four stud ies, two IATs on ir rel e vant di men sions (e.g., de li cious–happy) were sub stan tially cor re lated with IATs as s...