By Roland Ewald

To choose the main compatible simulation set of rules for a given job is frequently tricky. this can be as a result of problematic interactions among version beneficial properties, implementation information, and runtime atmosphere, which can strongly have an effect on the general functionality. an automatic choice of simulation algorithms helps clients in developing simulation experiments with no tough specialist wisdom on simulation.   Roland Ewald analyzes and discusses latest methods to resolve the set of rules choice challenge within the context of simulation. He introduces a framework for automated simulation set of rules choice and describes its integration into the open-source modelling and simulation framework James II. Its choice mechanisms may be able to take care of 3 events: no earlier wisdom is obtainable, the influence of challenge gains on simulator functionality is unknown, and a dating among challenge beneficial properties and set of rules functionality might be demonstrated empirically. the writer concludes with an experimental assessment of the built equipment.

Show description

Read or Download Automatic Algorithm Selection for Complex Simulation Problems PDF

Best nonfiction_4 books

Terpenoids and Steroids: Volume 5 (SPR Terpenoids and Steroids (RSC))

Professional Periodical studies supply systematic and distinctive evaluation assurance of development within the significant components of chemical study. Written via specialists of their professional fields the sequence creates a distinct carrier for the energetic learn chemist, providing general severe in-depth money owed of growth specifically components of chemistry.

Workbook to Accompany Fundamentals of Emergency Care

Basics of Emergency Care has a symptom orientated, review dependent method of the schooling of the EMT. it really is designed to learn utilizing a problem-based studying method. challenge established studying offers the EMT pupil a chance to profit the cloth in a simpler demeanour and person who is extra common to the grownup learner.

Additional resources for Automatic Algorithm Selection for Complex Simulation Problems

Example text

X∈P Problem Space Model, available resources. F Feature Extraction f (x) ∈ F = Rm Feature Space S( f (x), w) Model size, number of resources, etc. p(a, x) p ∈ Rn Performance Measure Space w ∈ Rn Criteria Space User preferences. a∈A Algorithm Space Speed, accuracy, memory consumption, etc. 3: ASP entities and exemplary correspondents from modeling and simulation. allows to compare their virtues and shortcomings on a rather abstract yet precise level. 2 Analytical Algorithm Selection A fundamental approach to assess problem hardness and thereby analytically compare the algorithms to solve them is provided by (computational) complexity theory.

X100 , say ||p(a1 , xi )|| = 10 versus ||p(a2 , xi )|| = 1 for i ∈ [1, 100]. Furthermore, let a2 perform only slightly better than a1 when applied to x101 , . . , x200 , say ||p(a1 , xi )|| = 10 and ||p(a2 , xi )|| = 11 for i ∈ [101, 200]. Now assume that solving the BSMP led to a selection mapping S that chooses a1 for x1 , . . , x70 , and a2 otherwise. S is average-effective since (see def. 15. This means that S is performing worse than a constant selection mapping that does not adapt its decision by considering any problem features.

23), this time on the grounds of a performance tuple set Φ: max φ1 ,φ2 ∈Φ∧( f φ1 ,aφ1 )=( f φ2 ,aφ2 ) ||pφ1 − pφ2 || Moreover, approximation theory can be applied to the given data in Φ: Is it possible to construct a good approximation function F(x, p1 , . . , pn ) → A (see eq. 11, p. 31)? , the deviation from the best possible algorithm selection? How to find the best parameters pi , and which approximation forms are suitable? 3 Algorithm Selection as Learning 37 foundation of two more practical disciplines [49]: statistical learning [122] and machine learning [333].

Download PDF sample

Automatic Algorithm Selection for Complex Simulation by Roland Ewald
Rated 4.87 of 5 – based on 18 votes