Iskanje po katalogu digitalne knjižnice

Opis: This paper defines blind spots in continuous optimization problems as global optima that are inherently difficult to locate due to deceptive, misleading, or barren regions in the fitness landscape. Such regions can mislead the search process, trap metaheuristic algorithms (MAs) in local optima, or hide global optima in isolated regions, making effective exploration particularly challenging. To address the issue of premature convergence caused by blind spots, we propose LTMA+ (Long-Term Memory Assistance Plus), a novel meta-approach that enhances the search capabilities of MAs. LTMA+ extends the original Long-Term Memory Assistance (LTMA) by introducing strategies for handling duplicate evaluations, shifting the search away from over-exploited regions and dynamically toward unexplored areas and thereby improving global search efficiency and robustness. We introduce the Blind Spot benchmark, a specialized test suite designed to expose weaknesses in exploration by embedding global optima within deceptive fitness landscapes. To validate LTMA+, we benchmark it against a diverse set of MAs selected from the EARS framework, chosen for their different exploration mechanisms and relevance to continuous optimization problems. The tested MAs include ABC, LSHADE, jDElscop, and the more recent GAOA and MRFO. The experimental results show that LTMA+ improves the success rates for all the tested MAs on the Blind Spot benchmark statistically significantly, enhances solution accuracy, and accelerates convergence to the global optima compared to standard MAs with and without LTMA. Furthermore, evaluations on standard benchmarks without blind spots, such as CEC’15 and the soil model problem, confirm that LTMA+ maintains strong optimization performance without introducing significant computational overhead.
Ključne besede: optimization, metaheuristics algorithm, algorithmic performance, duplicate solutions, nonrevisited solutions, blind spots, LTMA
Objavljeno v DKUM: 19.05.2025; Ogledov: 0; Prenosov: 2
Celotno besedilo (4,16 MB)

Opis: Reporting the empirical results of swarm and evolutionary computation algorithms is a challenging task with many possible difficulties. These difficulties stem from the stochastic nature of such algorithms, as well as their inability to guarantee an optimal solution in polynomial time. This research deals with measuring the performance of stochastic optimization algorithms, as well as the confidence intervals of the empirically obtained statistics. Traditionally, the arithmetic mean is used for measuring average performance, but we propose quantiles for measuring average, peak and bad-case performance, and give their interpretations in a relevant context for measuring the performance of the metaheuristics. In order to investigate the differences between arithmetic mean and quantiles, and to confirm possible benefits, we conducted experiments with 7 stochastic algorithms and 20 unconstrained continuous variable optimization problems. The experiments showed that median was a better measure of average performance than arithmetic mean, based on the observed solution quality. Out of 20 problem instances, a discrepancy between the arithmetic mean and median happened in 6 instances, out of which 5 were resolved in favor of median and 1 instance remained unresolved as a near tie. The arithmetic mean was completely inadequate for measuring average performance based on the observed number of function evaluations, while the 0.5 quantile (median) was suitable for that task. The quantiles also showed to be adequate for assessing peak performance and bad-case performance. In this paper, we also proposed a bootstrap method to calculate the confidence intervals of the probability of the empirically obtained quantiles. Considering the many advantages of using quantiles, including the ability to calculate probabilities of success in the case of multiple executions of the algorithm and the practically useful method of calculating confidence intervals, we recommend quantiles as the standard measure of peak, average and bad-case performance of stochastic optimization algorithms.
Ključne besede: algorithmic performance, experimental evaluation, metaheuristics, quantile, confidence interval, stochastic algorithms, evolutionary computation, swarm intelligence, experimental methodology
Objavljeno v DKUM: 28.03.2025; Ogledov: 0; Prenosov: 8
Celotno besedilo (490,48 KB)
Gradivo ima več datotek! Več...

Opis: Evolutionary algorithms have been shown to be very effective in solving complex optimization problems. This has driven the research community in the development of novel, even more efficient evolutionary algorithms. The newly proposed algorithms need to be evaluated and compared with existing state-of-the-art algorithms, usually by employing benchmarks. However, comparing evolutionary algorithms is a complicated task, which involves many factors that must be considered to ensure a fair and unbiased comparison. In this paper, we focus on the impact of stopping criteria in the comparison process. Their job is to stop the algorithms in such a way that each algorithm has a fair opportunity to solve the problem. Although they are not given much attention, they play a vital role in the comparison process. In the paper, we compared different stopping criteria with different settings, to show their impact on the comparison results. The results show that stopping criteria play a vital role in the comparison, as they can produce statistically significant differences in the rankings of evolutionary algorithms. The experiments have shown that in one case an algorithm consumed 50 times more evaluations in a single generation, giving it a considerable advantage when max gen was used as the stopping criterion, which puts the validity of most published work in question.
Ključne besede: evolutionary algorithms, stopping criteria, benchmarking, algorithm termination, algorithm comparison
Objavljeno v DKUM: 28.03.2025; Ogledov: 0; Prenosov: 2
Celotno besedilo (1,40 MB)
Gradivo ima več datotek! Več...

Opis: Namen diplomske naloge je bilo razviti 2D računalniško igro zvrsti obramba stolpov. V teoretičnem delu smo najprej opisali zvrsti računalniških iger obramba stolpov, opisali pomembnejše komponente, opisali dva starejša in dva bolj znana primera teh računalniških zvrsti in predstavili Unity računalniški pogon, v katerem smo razvijali igro. V praktičnem delu smo se osredotočili na sam razvoj igre, kjer smo najprej pripravili prazen projekt in vanj uvozili vsa potrebna sredstva. Razvili in opisali smo posamezne komponente igre in na koncu še implementirali uporabniški vmesnik.
Ključne besede: Razvoj računalniške igre, obramba stolpov, Unity, 2D
Objavljeno v DKUM: 08.08.2024; Ogledov: 290; Prenosov: 80
Celotno besedilo (3,20 MB)

Opis: To magistrsko delo je namenjeno implementaciji generatorja računalniških animacij rokopisa. Vsebuje opis podobnih obstoječih programov, kratek opis uporabljenih orodij, predvsem pa natančen opis spopadanja z različnimi izzivi med implementacijo. Unikatna lastnost našega generatorja rokopisa je to, da podpira uvoz lastnega rokopisa, ki ga pomožni program pridobi iz uporabnikovih videoposnetkov. Poleg generatorja rokopisa smo v tem delu ustvarili tudi omenjeno pomožno orodje za pridobivanje samega rokopisa iz uporabnikovih videoposnetkov. To smo naredili s pomočjo računanja delte med dvema sosednjima sličicama v videoposnetku, kar pa nam je z dokaj veliko natančnostjo omogočalo pridobivanje le vsebine, ki nas na videoposnetku zanima (animacijo rokopisa).
Ključne besede: Generiranje, animacija, predvajanje, posnetek, rokopis
Objavljeno v DKUM: 08.08.2024; Ogledov: 115; Prenosov: 41
Celotno besedilo (2,81 MB)