Wyniki wyszukiwania - BazTech

1

Different Coevolutionary Breeding Strategies of Ischnoceran Lice on Prunella collaris and P. modularis in High Mountains

Janiga M.

Polish Journal of Ecology

|

2018

|

Vol. 66, nr 2

182--193

EN

Alpine accentors (Prunella collaris) and dunnocks (P. modularis) are closely related species of Prunellidae, which often breed in polygynandrous groups and have specific types of mating behaviour. The alpine accentor lives at high altitudes, in an extreme alpine environment; the dunnock is widely dispersed from lowland to tree line in the mountains. Both species are hosts of the related species of wing lice Philoptersus emiliae and Ph. modularis, respectively. Behavioural differences between these two host species may have resulted in different breeding and seasonal adaptations by their parasites. The main goal of this paper was to test this hypothesis. Sixty five alpine accentors Prunella collaris (Scopoli, 1769) and eighty four dunnocks Prunella modularis (Linnaeus, 1758) were examined for Ischnocera lice in the West Carpathians, Slovakia from 1988 to 2001, and from 2007 to 2010, respectively. Birds were found to be infested with two species of Ischnoceran lice: the parasite species are not competing as they each have their own host Philopterus emiliae Balát, 1955 /P. collaris and Philopterus modularis (Denny, 1842) /P. modularis. Significant differences in abundance and prevalence existed between these two species of lice. High number of Ph. modularis nymphs in the comparison to Ph. emiliae nymphs reflects the different ecological, behavioural and phylogenetic trajectories of their host bird species. This work is the first to assess the louse breeding strategies on two closely related bird species in the high mountain environment.

2

Optymalizacja nieciągłych funkcji wielomodalnych z wykorzystaniem kooperacyjnego algorytmu koewolucyjnego

Kuczkowski Ł., Śmierzchalski R.

Zeszyty Naukowe Wydziału Elektrotechniki i Automatyki Politechniki Gdańskiej

|

2015

|

Nr 47

103--106

PL

W artykule przedstawiono algorytmy koewolucyjne, heurystyczną metodę rozwiązywania złożonych obliczeniowo problemów opartą na zasadzie korelacji oraz darwinowskiej teorii ewolucji. Opisano zalety algorytmu, możliwe zastosowania, sposób działania oraz niektóre z dotychczasowych implementacji. Następnie wybrano trzy wielomodalne lub nieciągłe funkcje testowe: Rosenbrocka, Styblinskiego-Tanga oraz Schaffer’a. Dokonano dekompozycji problemu wyznaczenia minimum globalnego funkcji i przeprowadzono optymalizację wykorzystując kooperacyjny algorytm koewolucyjny. Uzyskane wyniki pozwoliły na ocenę jakości działania algorytmu. Przeprowadzone testy i ich rezultaty są wstępem do szerszych badań nad algorytmami koewolucyjnymi.

EN

In this paper a brief study of coevolutionary algorithm is presented. The coevolutionary algorithm (CA) is an evolutionary algorithm (or collection of evolutionary algorithms) in which the fitness of an individual depends on the relationship between that individual and other individuals. CA can be divided into two fundamental sub-types. In cooperative algorithms, individuals are rewarded when they work well with other individuals and punished when they perform poorly. In competitive algorithms, however, individuals are rewarded at the expense of those with which they interact. The principle of operation of CA is quite similar to traditional evolutionary algorithm. The main deference lies in a fact that CA operate on multi-populations and evaluate individual based on its collaboration with individuals (collaborators) from other populations. Applying CA requires decomposition of the problem into components and assigning each component to a population. This article presents an optimization of discontinuous and multimodal functions using cooperative coevolutionry algorithm. The modified testing functions: Rosenbrocka, Styblinskiego-Tanga and Schaffer’a are decomposed and minimize using coevolutionary algorithm. Obtained results allow to evaluate the quality of the algorithm and will be used for further research on the topic.

3

Evolving small-board Go players using coevolutionary temporal difference learning with archives

Krawiec K., Jaśkowski W., Szubert M.

International Journal of Applied Mathematics and Computer Science

|

2011

|

Vol. 21, no 4

717-731

EN

We apply Coevolutionary Temporal Difference Learning (CTDL) to learn small-board Go strategies represented as weighted piece counters. CTDL is a randomized learning technique which interweaves two search processes that operate in the intra-game and inter-game mode. Intra-game learning is driven by gradient-descent Temporal Difference Learning (TDL), a reinforcement learning method that updates the board evaluation function according to differences observed between its values for consecutively visited game states. For the inter-game learning component, we provide a coevolutionary algorithm that maintains a sample of strategies and uses the outcomes of games played between them to iteratively modify the probability distribution, according to which new strategies are generated and added to the sample. We analyze CTDL's sensitivity to all important parameters, including the trace decay constant that controls the lookahead horizon of TDL, and the relative intensity of intra-game and inter-game learning. We also investigate how the presence of memory (an archive) affects the search performance, and find out that the archived approach is superior to other techniques considered here and produces strategies that outperform a handcrafted weighted piece counter strategy and simple liberty-based heuristics. This encouraging result can be potentially generalized not only to other strategy representations used for small-board Go, but also to various games and a broader class of problems, because CTDL is generic and does not rely on any problem-specific knowledge.