Focused Inference
نویسندگان
چکیده
We develop a method similar to variable elimination for computing approximate marginals in graphical models. An underlying notion in this method is that it is not always necessary to compute marginals over all the variables in the graph, but focus on a few variables of interest. The Focused Inference (FI) algorithm introduced reduces the original distribution to a simpler one over the variables of interest. This is done in an iterative manner where in each step the operations are guided by (local) optimality properties. We exemplify various properties of the focused inference algorithm and compare it with other methods. Numerical simulation indicates that FI outperform competing methods.
منابع مشابه
Improving Mobile Grid Performance Using Fuzzy Job Replica Count Determiner
Grid computing is a term referring to the combination of computer resources from multiple administrative domains to reach a common computational platform. Mobile Computing is a Generic word that introduces using of movable, handheld devices with wireless communication, for processing data. Mobile Computing focused on providing access to data, information, services and communications anywhere an...
متن کاملImproving Mobile Grid Performance Using Fuzzy Job Replica Count Determiner
Grid computing is a term referring to the combination of computer resources from multiple administrative domains to reach a common computational platform. Mobile Computing is a Generic word that introduces using of movable, handheld devices with wireless communication, for processing data. Mobile Computing focused on providing access to data, information, services and communications anywhere an...
متن کاملThe 'Ah Ha!' Moment: When Possible, Answering the Currently Unanswerable using Focused Reasoning
Focused reasoning is a method for continuing a specific inference task as soon as rules or facts which may assist in the reasoning are added to the knowledge base without repeating completed inference, re-posing queries, or performing unnecessary inference. Determining if focused reasoning should commence uses very few computational resources above those used normally to add a term to a knowled...
متن کاملFocused Grounding for Markov Logic Networks
Markov logic networks have been successfully applied to many problems in AI. However, the computational complexity of the inference procedures has limited their application. Previous work in lifted inference, lazy inference and cutting plane inference has identified cases where the entire ground network need not be constructed. These approaches are specific to particular inference procedures, a...
متن کاملDesign On-Line Tunable Gain Artificial Nonlinear Controller
One of the most important challenges in nonlinear, multi-input multi-output (MIMO) and time variant systems (e.g., robot manipulator) is designing a controller with acceptable performance. This paper focused on design a new artificial non linear controller with on line tunable gain applied in the robot manipulator. The sliding mode fuzzy controller (SMFC) was designed as 7 rules Mamdani’s infer...
متن کاملDesign On-Line Tunable Gain Artificial Nonlinear Controller
One of the most important challenges in nonlinear, multi-input multi-output (MIMO) and time variant systems (e.g., robot manipulator) is designing a controller with acceptable performance. This paper focused on design a new artificial non linear controller with on line tunable gain applied in the robot manipulator. The sliding mode fuzzy controller (SMFC) was designed as 7 rules Mamdani’s infer...
متن کامل