|
|
Using reinforcement learning to learn how to play text-based games
Zelinka, Mikuláš ; Kadlec, Rudolf (advisor)
The ability to learn optimal control policies in systems where action space is defined by sentences in natural language would allow many interesting real-world applications such as automatic optimisation of dialogue systems. Text-based games with multiple endings and rewards are a promising platform for this task, since their feedback allows us to employ reinforcement learning techniques to jointly learn text representations and control policies. We present a general text game playing agent, testing its generalisation and transfer learning performance and showing its ability to play multiple games at once. We also present pyfiction, an open-source library for universal access to different text games that could, together with our agent that implements its interface, serve as a baseline for future research.
|
|
|
Adaptive agents and emotions
Kadlec, Rudolf ; Brom, Cyril (advisor) ; Kryl, Rudolf (referee)
This thesis investigates possible assets of emotions for autonomous adaptive agents working in environments similar to the real world. In living organisms, emotions have developed as a mechanism of adaptation to the surrounding environment. Therefore it is worth asking whether mechanisms similar to emotions can be implemented in models of autonomous agents. In this thesis a model of ethology inspired agent using reinforcement learning was implemented. This model suggests that emotions influence the balance between exploring new strategies and exploiting the strategies already known (the so-called explore/exploit problem). Negative emotional evaluation leads to changes in action selection strategy. The emotional version proved to be better than the non-emotional one in some environments. In other types of environments, the expectations have not been fulfilled. The instability of the received results is probably caused by non-optimal parameterization of the whole model.
|
|
|
Evoluce chování inteligentních agentů v počítačových hrách
Kadlec, Rudolf ; Kudová, Petra (advisor) ; Holan, Tomáš (referee)
In the present work we study evolution of both high-level and low-level behaviour of agents in the environment of the commercial game Unreal Tournament 2004. For optimization of high-level behaviour in Deathmatch and Capture the flag game modes a new functional architecture for description of player's behaviour was designed and implemented. Then a genetic programming technique was used to optimise it. Experiments with both standard evolution schema and with coevolution are presented. In second series of experiments the NEAT algo- rithm was used to evolve low-level missile avoidance behaviour (so called "dodging").
|
|
|
Generating life episodes for the purpose of testing of episodic memory models
Běhan, Zdeněk ; Kadlec, Rudolf (advisor) ; Bída, Michal (referee)
The goal of this work is to create a generator that provides a corpora of input episodes in the specified format, which can be used as an input to test episodic memory models. More specifically, the methods used should ensure the scope to be in years up to a possible lifetime of a typical human agent. It also attempts to verify this on an actual episodic memory model, and test if the generated data has high enough quality to be used for testing psychological paradigms on memory models.
|
|
|
Evolution of the use of memory
Polák, Marek ; Holan, Tomáš (advisor) ; Kadlec, Rudolf (referee)
The task of the bachelor work is to give a proposal and its implementation of application for simulating evolution of population. The outcome is a study of dependency of memory usage on energy of organishms in their environment. From ndings of simulation batches it could be said what parameters are useful for given environment from view of count of conditions to spent of energy, which organishm leeds to its life.
|
|
|
Episodic Memory with Believable Forgetting
Soukup, Tomáš ; Brom, Cyril (advisor) ; Kadlec, Rudolf (referee)
Title: Episodic Memory with Believable Forgetting Author: Tomáš Soukup Department: Department of Software and Computer Science Education Supervisor: Mgr. Cyril Brom, Ph.D., DSCSE Abstract: Presented thesis introduces a model of episodic memory for virtual humans extended by believable forgetting based on rating of memories according to their importance. It is inspired by a psychological model of E. Tulving and experiments of W.A. Wagenaar and M. Linton and builds on a memory model for a human-like agent developed by Klára Pešková, which we modified for the needs of forgetting. Our model takes advantage of the level-of-detail approach to forget the parts of memories gradually. In addition to the age and particularity of memories, emotiveness is also considered during the rating of memories. A simple emotional model was created for this purpose. The functionality of our model was verified by implementing it into a prototype application, which simulates the life of a virtual human in a virtual world. Our experiments showed that the behavior of our model, when configured properly, corresponds with the psychological concepts. Keywords: episodical memory, forgetting, virtual human, emotion, believability
|
|
|
Integration of process modelling into the multiagent framework IVE
Návrat, Jan ; Brom, Cyril (advisor) ; Kadlec, Rudolf (referee)
The thesis focuses on a new method of designing and controlling multiagent worlds, applying the technique named "process modeling" currently used by commercial companies. The goal is to use the process procedures in order to improve and simplify designing of multiagent worlds. The thesis describes the design and implementation of a tool for process modeling extended by the possibility of IVE multiagent world simulation. The IVE multiagent world is further expanded by adding the tools for monitoring the extended processes and the possibility to use them in order to control these worlds. The thesis finally summarizes the main advantages of the new method and outlines possible further improvements. The reader of the thesis is also introduced to the area of agents, multiagent worlds and process modeling.
|
|
|
DyBaNeM: Bayesian Model of Episodic Memory
Kadlec, Rudolf ; Brom, Cyril (advisor) ; Lim, Mei Yii (referee) ; Pilát, Martin (referee)
Title: DyBaNeM: Bayesian Model of Episodic Memory Author: Mgr. Rudolf Kadlec E-mail: rudolf.kadlec@gmail.com Department: Department of Software and Computer Science Education Supervisor: Mgr. Cyril Brom, Ph.D. Department of Software and Computer Science Education Abstract: Artificial agents endowed with episodic (or autobiographic) memory systems have the abilities to remember and recall what happened to them in the past. The existing Episodic Memory (EM) models work as mere data-logs with indexes: they enable record, retrieval and delete operations, but rarely organize events in a hierarchical fashion, let alone abstract automatically detailed streams of "what has just happened" to a "gist of the episode." Consequently, the most interest- ing features of human EM, reconstructive memory retrieval, emergence of false memory phenomena, gradual forgetting and predicting surprising situations are out of their reach. In this work we introduce a computational framework for episodic memory modeling called DyBaNeM. DyBaNeM connects episodic mem- ory abilities and activity recognition algorithms and unites these two computer science themes in one framework. This framework can be conceived as a general architecture of episodic memory systems, it capitalizes on Bayesian statistics and, from the psychological...
|
|
|
Coordination of multiple virtual agents in team-based computer games
Kolombo, Martin ; Gemrot, Jakub (advisor) ; Kadlec, Rudolf (referee)
Práce se zabývá problémem efektivní koordinace týmu virtuálních agent· v počí- tačových hrách. Hlavním cílem bylo navrhnou obecnou architektru pro práci s informacemi o virtuálním prostředí, která poskytuje programátorovi dobrý přístup k týmovým znalostem o virtuálním prostředí. Zvolená architektura neklade na programátora žádné požadavky ohledně implementace týmu agent· a nezávisí na žádném konkrétním virtuálním prostředí. Architektura je založena na přirozeném rozdělení dat podle jejich schopnosti se měnit v pr·běhu simulace a podle jejich subjektivity. Tato distribuce dat přirozeně vytváří sdílené znalosti týmu a programátorovi tak umožní založit rozhodování agent· na znalostech celého týmu namísto pouze jediného agenta. Architekturu jsme implemento- vali do systému Pogamut, ve kterém jsme provedli validaci zpětné kompati- bility. Vzheldem k nekompatibilitě formátu dat exportovaných součástí Poga- mutu, který jsme objevili v poslední fázi validace jsme provedli validaci pouze částečnou. Úpravu platformy Pogamut a následnou implementaci týmu agent· využívajícího naši architekturu plánujeme jako pokračování práce. 1
|
|
|
Suffix tree construction with minimized branching
Bašista, Peter ; Dvořák, Tomáš (advisor) ; Kadlec, Rudolf (referee)
Suffix tree is a data structure which enables the performing of fast search-like operations on the text. In order to be used efficiently, it must be created quickly. In this thesis, we focus on the new kind of suffix link simulation called "minimized branching", which aims to increase the speed of suffix tree construction by reducing the number of branching operations. Our main goal is to present a comparison of the currently used methods for the suffix tree construction and to point out some advantages and disadvantages of the individual approaches. We introduce, implement and practically evaluate multiple variations of the standard McCreight's and Ukkonen's algorithms, as well as Partition and Write Only Top Down (PWOTD) algorithm, originally developed for disk-based construction. Our main result is the integrated description and implementation of these algorithms, which are both well-suited to be further built upon. We also present a simple recommendations on when it is advisable to use a particular algorithm's variation and why.
|
guest ::
RSS feed.