|
|
Coevolution of AI and level generation for Super Mario game
Flimmel, Július ; Černý, Vojtěch (advisor) ; Pilát, Martin (referee)
Procedural Content Generation is now used in many games to generate a wide variety of content. It often uses players controlled by Artificial Intelligence for its evaluation. PCG content can also be used when training AI players to achieve better generalization. In both of these fields, evolutionary algorithms are employed, but they are rarely used together. In this thesis, we use the coevolution of AI players and level generators for platformer game Super Mario. Coevolution's benefit is, that the AI players are evaluated by adapting level generators, and vice versa, level generators are evaluated by adapting AI players. This approach has two results. The first one is a creation of multiple level generators, each generating levels of gradually increased difficulty. Levels generated using a sequence of these generators also mirror the learning curve of the AI player. This can be useful also for human players playing the game for the first time. The second result is an AI player, which was evolved on gradually more difficult levels. Making it learn progressively may yield better results. Using the coevolution also doesn't require any training data set.
|
|
|
Animal phobia: fear or disgust?
Peléšková, Šárka ; Landová, Eva (advisor) ; Hocko Fajnerová, Iveta (referee)
Animal phobias are among the most common specific phobias. It is traditionally assumed that two basic emotions are involved - fear and anxiety, although recent studies suggest that disgust may also be important. Phobias and involved emotions are defined mostly by psychologists and psychiatrists. Their manifestations could be identified mainly on physiological level and are being researched using advanced medical methods (e.g. functional neuroimaging). The eliciting stimuli of arachnophobia, ophidiophobia etc. are specific animal, spiders or snakes. However the category of elicitors for disgust sensu lato is fairly extended. The etiology of these phenomena is explained from the perspective of evolutionary biology and presumes biologicaly significant selective pressures and common coevolution. The aim of this paper is to summarize information gathered so far, verify proposed evolutionary hypothesis about the involvement of disgust in the etiology of animal phobia and to evaluate biological relevancy of fear of animal stimuli.
|
|
|
Phylogeny of selected species groups of the genus Torymus (Hymenoptera: Torymidae)
Křížková, Barbora ; Janšta, Petr (advisor) ; Skuhrovec, Jiří (referee)
Chalcid wasps (Chalciodiodea) is superfamily within parasitic group of the order Hymenoptera. These insects attack other arthropods. My thesis is focused on phylogeny of selected species groups of the genus Torymus Dalman 1820, whose range of hosts includes mainly gall-forming insect as gall wasps (Hymenoptera: Cynipidae) and gall midges (Diptera: Cecidomyiidae). Using genes 28D2 rDNA, COI and CytB, I constructed phylogenetic trees, which helped with discussing current view on classification of the genus and with inferring new findings about co-evolution with host organisms. Previously established morphological - ecological species groups were not supported by my analyses in many cases. Morphological traits seem to be convergent for many species. They often do not support observed monophyla as apomorphies. On the other hand, ecology and natural habitat of species was common for many branches in my cladograms. It was found out, that recent hosts are not probably original hosts of the group. Adaptive radiation in different habitats is likely to be the phenomenon that stands behind present-day host range of Torymus species. This process could have occurred after swapping to gall midges and gall wasps. Key words: Chalcidoidea, Torymus, phylogeny, parasitoid, host specifity, co-evolution
|
|
|
Preferences of the main groups of pollinators for the functional floral traits
Malá, Karolína ; Janovský, Zdeněk (advisor) ; Hadrava, Jiří (referee)
This bachelor thesis summarizes the preferences of the main functional groups of pollinators towards floral traits of plants. Due to coevolution and mutual selection pressures, these two life forms have begun to adapt to each other and develop traits on which the future of their life depends. The plant seeks to attract its pollinators, thus ensuring efficient pollen transfer or reproduction for its species. Attractants are used to motivate visitors to its flowers. These attractants provide pollinators with either direct rewards (food, shelter, brood) or rewards in the form of sensory perceptions (colour, scent, morphology). Recently, it has become apparent that flower colour, morphology and the type of food reward offered by the flower have the greatest influence on pollinator choice. Based on preferences towards attractants, we can organize several functional groups whose representatives have the same or similar choice preferences. Key words Pollination syndromes, pollinator, selection pressures, plant traits, coevolution, preferences, fidelity, diptera, coleoptera, lepidoptera, hymenoptera, birds, colour, morphology, rewards
|
|
|
Phylogeny of selected species groups of the genus Torymus (Hymenoptera: Torymidae)
Křížková, Barbora
Chalcid wasps (Chalciodiodea) is superfamily within parasitic group of the order Hymenoptera. These insects attack other arthropods. My thesis is focused on phylogeny of selected species groups of the genus Torymus Dalman 1820, whose range of hosts includes mainly gall-forming insect as gall wasps (Hymenoptera: Cynipidae) and gall midges (Diptera: Cecidomyiidae). Using genes 28D2 rDNA, COI and CytB, I constructed phylogenetic trees, which helped with discussing current view on classification of the genus and with inferring new findings about co-evolution with host organisms. Previously established morphological - ecological species groups were not supported by my analyses in many cases. Morphological traits seem to be convergent for many species. They often do not support observed monophyla as apomorphies. On the other hand, ecology and natural habitat of species was common for many branches in my cladograms. It was found out, that recent hosts are not probably original hosts of the group. Adaptive radiation in different habitats is likely to be the phenomenon that stands behind present-day host range of Torymus species. This process could have occurred after swapping to gall midges and gall wasps. Key words: Chalcidoidea, Torymus, phylogeny, parasitoid, host specifity, co-evolution
|
|
|
Coevolution of AI and level generation for Super Mario game
Flimmel, Július ; Černý, Vojtěch (advisor) ; Pilát, Martin (referee)
Procedural Content Generation is now used in many games to generate a wide variety of content. It often uses players controlled by Artificial Intelligence for its evaluation. PCG content can also be used when training AI players to achieve better generalization. In both of these fields, evolutionary algorithms are employed, but they are rarely used together. In this thesis, we use the coevolution of AI players and level generators for platformer game Super Mario. Coevolution's benefit is, that the AI players are evaluated by adapting level generators, and vice versa, level generators are evaluated by adapting AI players. This approach has two results. The first one is a creation of multiple level generators, each generating levels of gradually increased difficulty. Levels generated using a sequence of these generators also mirror the learning curve of the AI player. This can be useful also for human players playing the game for the first time. The second result is an AI player, which was evolved on gradually more difficult levels. Making it learn progressively may yield better results. Using the coevolution also doesn't require any training data set.
|
|
|
Coevolution of avian ectoparasites in the tropics
Gajdošová, Magdalena ; Munclinger, Pavel (advisor) ; Votýpka, Jan (referee)
Host-parasite associations are born by cospeciations or by host switches. Feather lice have traditionally been used as model parasites for studying these events and underlying ecological factors. By now tens of analyses have addressed comparisons of host and parasite phylogenies to study cospeciations and host switches in lice, however, these analyses are strongly biased towards the temperate zone. Tropical environment could provide new insight into the origin of host-parasite interactions, because it is ecologically unique. This work aims to supplement the knowledge of host-parasite associations in lice using coevolution analyses of two feather lice genera and their passerine hosts in tropical rainforest in Cameroon. It shows that lice in the tropics cospeciate rarely. To assess whether host switches are non-random and occur preferentially between hosts with specific traits, this work also analyses relations between parasite genetic distances and hosts' trait similarities. No effect of host morphology and spatial distribution was found. However, genetic distances of the lice strongly correlate with genetic distances of their hosts.
|
|
|
Procedural Generation of Endless Runner Type of Video Games
Černý, Vojtěch ; Gemrot, Jakub (advisor) ; Pilát, Martin (referee)
Procedural content generation (PCG) is increasingly used to generate many aspects in a variety of games. AI players, both hand scripted or also generated (by AI methods), are used to evaluate this content. Comparatively little effort is invested in using PCG to generate the whole game, including its rules. In this thesis, we use evolutionary algorithms to generate the game rules, its content and the evaluating AI player on a narrow, but flourishing, genre of endless runners - games where the player is constantly running. For this purpose, we have implemented a framework for creating endless runner games. Our approach could provide more efficiency for game designers, explore completely new game concepts in endless runners, platformer games, and be further generalized to other game genres.
|
|
|
Image Classification Using Genetic Programming
Jašíčková, Karolína ; Vašíček, Zdeněk (referee) ; Sekanina, Lukáš (advisor)
This thesis deals with image classification based on genetic programming and coevolution. Genetic programming algorithms make generating executable structures possible, which allows us to design solutions in form of programs. Using coevolution with the fitness prediction lowers the amount of time consumed by fitness evaluation and, therefore, also the execution time. The thesis describes a theoretical background of evolutionary algorithms and, in particular, cartesian genetic programming. We also describe coevolutionary algorithms properties and especially the proposed method for the image classifier evolution using coevolution of fitness predictors, where the objective is to find a good compromise between the classification accuracy, design time and classifier complexity. A part of the thesis is implementation of the proposed method, conducting the experiments and comparison of obtained results with other methods.
|
|
|
Bioinformatic methods of detection of protein coevolution
Pařízková, Hana ; Schneider, Bohdan (advisor) ; Hampl, Vladimír (referee)
The term coevolution describes the situation when two or more species or biomole- cules reciprocally affect each others' evolution. On the protein level, it is thought to be the main mechanism ensuring correct folding, interactions and function of a protein, and it can be observed both on the level of interacting protein families and individual amino acid residues. Coevolution studies have been proved to be a powerful tool for prediction of protein structure, function, interaction partners, etc. In this thesis, different algorithms used for detection of protein coevolution are described, as well as their applications and limitations. Keywords: coevolution, protein family, protein structure prediction, interac- tion partners, correlated mutations, mirrortree, mutual information, direct cou- pling analysis
|
guest ::
