690 research outputs found
Institutionalizing ASEAN: celebrating Europe through network governance
No full textThis article provides a new piece for two of the puzzles of institutionalized cooperation in Association of Southeast Asian Nations (ASEAN). First, with regard to the organization's four decades of existence, there has always been a marked gap between ASEAN's rhetorical goals of cooperation and its actual achievements. What explains these systematic failures of implementation? Second, from the outset, ASEAN was criticized for its light institutionalization, which failed to deliver the substantial cooperation goals. Despite selected institutional reforms, ASEAN's autonomy has not increased remarkably and it has not made any major institutional innovations. Why does ASEAN design institutions it does not use? Why does this transformation gap occur? The author suggests a sociological institutional explanation and argues that major impulses for cooperation have come from outside Southeast Asia, most importantly from Europe. By mimicking the European integration process, ASEAN member states have effectively created an isomorphic organization. The Association's institutional development reflects a concern for international legitimacy and less an objective functional demand arising from the specific interactions of member states. This copying process has led to network governance within the organization
The Vietnamese Theatrical Tradition in the Plot of Monsun, by Anja Hillin
No full textFrom the dramatic text Monsun by the German author Anja Hilling, this article analyzes the construction of a plot using elements of the Asian theater, in this specific case the Vietnamese tradition, identifying its presence inthe structure beyond the constant references to the context of that country, starting with the title of the play itself and the trip of one of the characters to Vietnam.A partir do texto dramático Monsun, da autora alemã Anja Hilling, este artigo analisa a construção de uma trama que se utiliza de elementos do teatro asiático, neste caso específico a tradição vietnamita, identi cando a sua presença na estrutura, além das men- ções constantes ao contexto daquele país, começando pelo próprio título da peça e pela viagem de uma das personagens ao Vietnam. 
Discussing Anja Štefan`s fairytales in the first year of primary school
No full textFairytales are an indispensable part in children's lives. They stimulate the development of imagination, fantasy; they help to enrich children’s vocabulary and with expressing feelings. There are different types of fairytales and it is very important that children get to know the variety of them. I firstly presented a youth author Anja Štefan and her literary works in my diploma paper. Later on I continued with the presentation of youth literature and I defined the theory of a fairytale, wrote about the origin, its meaning and classification of fairytales which I have also described in greater detail. I stated and described authorial prose work of Anja Štefan: picture book Lešniki, lešniki and classic fairytales Kotiček na koncu sveta, Melje, melje mlinček and Štiri črne mravljice, which can be defined as animal fairytales. I was interested in the receptive development of children in the period of intuitive intelligence, which I thoroughly described. In the empirical part of my diploma I wrote lesson plans for discussing and teaching three Anja Štefan’s animal fairytales in the first grade of primary school. 25 children took part in the research. I found out that pupils distinguish between real and unreal life (imaginative life), they are capable of identifying themselves with a literary person and they perceive animals as literary persons and they identify themselves with them
Machine learning algorithms applied to a given data set
No full textU svijetu podataka i velike količine informacija algoritmi umjetne inteligencije postaju sve značajniji. U ovisnosti o vrsti i strukturi podataka te metodi strojnog učenja algoritmi se koriste prilikom predviđanja, odlučivanja, klasifikacije, prepoznavanja i slično. U ovom radu opisane su i objašnjene tri klasifikacijske neparametarske metode strojnog učenja koje pripadaju nadgledanom učenju – stroj potpornih vektora, algoritam k-najbližih susjeda te stablo odluke. Sva tri algoritma su primijenjena na zadanom skupu podataka u programskom jeziku Python te open-source web aplikaciji Jupyter Notebook. Prije same primjene algoritama na zadani skup podataka, na podatke je bilo potrebno primijeniti funkcije obrade podataka – funkcije čitanja, funkcije čišćenja, funkcije skaliranja te funkcije vizualizacije. Algoritmi su uspoređeni po validacijskom kriteriju preciznosti.In the world of data and large amounts of information, artificial intelligence algorithms are becoming increasingly important. Depending on the type and structure of the data and the machine learning method, algorithms are used for prediction, decision-making, classification, recognition, and the like. In this paper, three classification non-parametric machine learning methods that belong to supervised learning are described and explained - the support vector machine, the k-nearest neighbors algorithm and the decision tree. All three algorithms were applied to the given data set in the Python programming language and the open-source web application Jupyter Notebook. Before applying the algorithms to the given data set, it was necessary to apply data processing functions to the data - reading functions, cleaning functions, scaling functions and visualization functions. Algorithms were compared according to the validation criterion of precision
Development of an intelligent agent based on deep learning
No full textU ovom diplomskom radu istražene su i primijenjene metode pojačanog i dubokog učenja u svrhu razvoja inteligentnog agenta sposobnog za učenje igranja videoigre kroz interakciju s okolinom. U teorijskom dijelu rada obrađeni su osnovni pojmovi pojačanog učenja (Reinforcement Learning), model Q-learning, te duboke neuronske mreže s naglaskom na konvolucijske neuronske mreže (CNN), koje su korištene zbog obrade vizualnih ulaznih podataka. U sklopu rješenja primijenjen je deep Q-network (DQN) model kao metoda učenja optimalne politike ponašanja iz vizualnog prikaza okoline.
Praktični dio rada obuhvaća razvoj simulacijskog okruženja temeljeno na jednoj od Atari igara, u kojem agent uči igranjem i prilagođava svoje ponašanje ovisno o povratnim informacijama iz okoline. Implementiran je DQN model u programskom jeziku Python, korištenjem PyCharm razvojne okoline. Razvijene su i testirane različite arhitekture neuronskih mreža, uz prilagodbu hiperparametara, dok je proces treniranja i evaluacije modela detaljno analiziran pomoću alata TensorBoard. Konačno, najbolji model pokrenut je u zasebnoj skripti za evaluaciju izvedbe agenta.This thesis explores and applies methods of reinforcement learning and deep learning with the goal of developing an intelligent agent capable of learning to play a video game through interaction with its environment. The theoretical part of the thesis covers fundamental concepts of reinforcement learning, the Q-learning algorithm, and deep neural networks, with a focus on convolutional neural networks (CNN), which were used due to their ability to process visual input data. As part of the implemented solution, the Deep Q-Network (DQN) model was used as a method for learning an optimal behavior policy based on the visual representation of the environment.
The practical part of the thesis includes the development of a simulation environment based on one of the Atari games, in which the agent learns by playing and adjusts its behavior based on feedback from the environment. The DQN model was implemented in the Python programming language using the PyCharm development environment. Various neural network architectures were developed and tested, with adjustments to hyperparameters, while the training and evaluation process was thoroughly analyzed using the TensorBoard tool. Finally, the best-performing model was executed in a separate script for agent performance evaluation
Correction to: Predicting the effectiveness of protected areas of Natura 2000 under climate change
No full textIn the original publication of this article (Nila & Hossain, 2019), co-authors ‘Carl Beierkuhnlein, Anja Jaeschke and Samuel Hoffmann’ need to be added to the author list
The establishment history panel 1975-2006 : handbook Version 2.0.1
No full text"The Establishment History Panel (BHP) for 1975 to 2006 includes all of the establishments throughout Germany which have at least one employee liable to social security as of the 30th of June of a given year or which have also had at least one marginal part-time employee as of 1999. For establishments in West Germany the observation period is the period 1975 to 2006 and for establishments in East Germany 1991 to 2006. The data source for the BHP is the Employee and Benefit Recipient History (BLH) of the IAB. The data on individuals contained in the BLH are aggregated to establishment level using the establishment numbers. The BHP covers between 1.3 and 2.7 million establishments per year. The individual annual waves can be linked to a panel dataset via the identifier 'artificial establishment number', which can be found in the dataset. This data report describes the variables of the weakly anonymous version of the 1975-2006 BHP basic dataset, which is available to researchers for on-site use at the Research Data Centre (FDZ) or for remote data access also. These BHP data are predominantly original data, with the sole exception that the original establishment numbers, which are direct identifiers for the different establishments, were replaced by artificial establishment numbers. The sensitive characteristics contained in the dataset, i.e. German Classification of Economic Activities 93 as 5- digit codes, German Classification of Economic Activities 03 as 5-digit codes, and Place of Work (District), are made available in the original on request and in well-founded cases only; all other users are given these variables in their aggregate forms only (as 3-digit codes or according to German States). This report will begin with a brief description of the sources used for the BHP, as well as the generation and processing of the BHP, followed by individual detailed explanations on the variables of the BHP 1975-2006. This data report contains no frequency tables for the individual variables. These are given in a separate text file that can be found on the FDZ homepage. To begin with, a brief overview of the changes to the BHP 1975-2006 in relation to the previous version from 1975-2005 is provided. This will help users of this version come to terms with the new version." (Author's abstract, IAB-Doku) ((en)) Additional Information Here you find frequency tables for the individual variables. Here you can find the German version.IAB-Betriebs-Historik-Panel, IAB-Beschäftigtenhistorik, IAB-Leistungsempfängerhistorik, Datenaufbereitung, Forschungsdatenzentrum, Arbeitsmarktforschung, Datenorganisation, Betrieb, Wirtschaftszweige, regionale Verteilung, Unternehmensgründung, Betriebsstilllegung, Beschäftigtenstruktur, sozialversicherungspflichtige Arbeitnehmer, erwerbstätige Frauen, Inländer, Ungelernte, Qualifikationsniveau, Bildungsabschluss, Schulbildung, Berufsausbildung, Studienabschluss, Fachhochschule, Hochschule, Vollzeitarbeit, Stellung im Beruf, Auszubildende, nicht formal Qualifizierte, Facharbeiter, Polier, Meister, Angestellte, Teilzeitarbeitnehmer, Heimarbeitnehmer, geringfügige Beschäftigung, Altersstruktur, Lohnstruktur, Lohnhöhe, Vollzeitarbeit, Nationalität, Jugoslawe, Osteuropäer, Pole, Spanier, Portugiese, Türke, Beschäftigtenstatistik
The establishment history panel 1975-2005 : handbook version 2.0.0
No full text"The Establishment History Panel (BHP) for 1975 to 2005 includes all of the establishments throughout Germany which have at least one employee liable to social security as of the 30th of June of a given year or which have also had at least one marginal part-time employee as of 1999. For establishments in West Germany the observation period is the period 1975 to 2005 and for establishments in East Germany 1991 to 2005. The data source for the BHP is the Employee and Benefit Recipient History (BLH) of the IAB. The data on individuals contained in the BLH are aggregated to establishment level using the establishment numbers. The BHP covers between 1.3 and 2.5 million establishments per year. This data report describes the variables of the weakly anonymous version of the 1975-2005 BHP basic dataset, which is available to researchers for on-site use at the Research Data Centre (FDZ) only. These BHP data are predominantly original data, with the sole exception that the original establishment numbers, which are direct identifiers for the different establishments, were replaced by artificial establishment numbers. The sensitive characteristics contained in the dataset, i.e. German Classification of Economic Activities 93 as 5-digit codes, German Classification of Economic Activities 03 as 5-digit codes, and Place of Work (District), are made available in the original on request and in well-founded cases only; all other users are given these variables in their aggregate forms only (as 3-digit codes or according to German States). This report will begin with a brief description of the sources used for the BHP, as well as the generation and processing of the BHP, followed by individual detailed explanations on the variables of the BHP 1975-2005. This data report contains no frequency tables for the individual variables. These are given in a separate text file that can be found on the FDZ homepage. To begin with, a brief overview of the changes to the BHP 1975-2005 in relation to the previous version from 1993-2003 is provided. This will help users of this version come to terms with the new version." (Author's abstract, IAB-Doku) ((en)) Additional Information Here you can find the german version of the documentation.IAB-Betriebs-Historik-Panel, IAB-Beschäftigtenhistorik, IAB-Leistungsempfängerhistorik, Datenaufbereitung, Forschungsdatenzentrum, Arbeitsmarktforschung, Datenorganisation, Betrieb, Wirtschaftszweige, regionale Verteilung, Unternehmensgründung, Betriebsstilllegung, Beschäftigtenstruktur, sozialversicherungspflichtige Arbeitnehmer, erwerbstätige Frauen, Inländer, Ungelernte, Qualifikationsniveau, Bildungsabschluss, Schulbildung, Berufsausbildung, Studienabschluss, Fachhochschule, Hochschule, Vollzeitarbeit, Stellung im Beruf, Auszubildende, nicht formal Qualifizierte, Facharbeiter, Polier, Meister, Angestellte, Teilzeitarbeitnehmer, Heimarbeitnehmer, geringfügige Beschäftigung, Altersstruktur, Lohnstruktur, Lohnhöhe, Vollzeitarbeit, Nationalität, Jugoslawe, Osteuropäer, Pole, Spanier, Portugiese, Türke, Beschäftigtenstatistik
- …
