3 results
Search Results
Now showing 1 - 3 of 3
Article Citation - WoS: 4Citation - Scopus: 6An Automata Networks Based Preprocessing Technique for Artificial Neural Network Modelling of Primary Production Levels in Reservoirs(Elsevier, 2007) Kilic, Hurevren; Soyupak, Selcuk; Tuzun, Ilhami; Ince, Ozlem; Basaran, GokbenPrimary production in lakes and reservoirs develops as a result of complex reactions and interactions. Artificial neural networks (ANN) emerges as an approach in quantification of primary productivity in reservoirs. Almost all of the past ANN applications employed input data matrices whose vectors represent either water quality parameters or environmental characteristics. Most of the time, the components of input matrices are determined using expert opinion that implies possible factors that affect output vector. Major disadvantage of this approach is the possibility of ending-up with an input matrix that may have high correlations between some of its components. In this paper, an automata networks (AN) based preprocessing technique was developed to select suitable and appropriate constituents of input matrix to eliminate redundancy and to enhance calculation efficiency. The proposed technique specifically provides an apriori rough behavioral modeling through identification of minimal AN interaction topology. Predictive ANN models of primary production levels were developed for a reservoir following AN based pre-modeling step. The achieved levels of model precisions and performances were acceptable: the calculated root mean square error values (RMSE) were low; a correlation coefficient (R) as high as 0.83 was achieved with an ANN model of a specific structure. (c) 2006 Elsevier B.V. All rights reserved.Article Citation - WoS: 1Citation - Scopus: 2Promotion of Cooperation in a Co-Evolutionary Pragmatic Agent Multigame Environment(Elsevier, 2025) Kilic, Omer Durukan; Kilic, HurevrenThe promotion of cooperation in a co-evolutionary environment where pragmatic agents participate in a multigame setting that contains Prisoner's Dilemma (PD) and SnowDrift (SD) games is investigated. The pragmatic agent conserves its current perspective when successful; otherwise adopts the opposite perspective. Unlike traditional models, this study introduces a setup in which perception and strategy spaces co-evolve in terms of iterative game payoffs. The players are situated in a 2-D square lattice environment and synchronously update their perceptions and strategies after interacting with their immediate neighbors. The ratios of perceptions and strategy are randomly set based on parameters alpha and /3, indicating initial SD and cooperation (C) percentages, respectively. By the end of various simulations, the system's convergent and stable behavior is shown by means, standard deviations, and confidence intervals. The results show that larger (alpha >= 0.5) initial populations of SD agents promote greater cooperation and lead to a dominance of cooperative strategies even for smaller initial C strategies (/3 similar to 0.2). Conversely, when the environment is initially dominated by PD perspectives and defect (D) strategies (alpha = 0.1, /3 = 0.2), it leads to lower levels of cooperation. Depending on the initial ratios of PD/SD and D/C players, cooperative player clusters are not only formed but are also persistent parts of the environment. Finally, we observed that co-evolving PD/SD and D/C environments coupled with pragmatic players lead to a controllable promotion of cooperation even against small initial SD player ratios.Article Citation - Scopus: 1A Player Reputation System Based on Belief Formation Among Non-Player Character Societies in Open-World Role-Playing Games(Elsevier Sci Ltd, 2023) Aydin, Ali Murat; Kilic, Hurevren; Guran, AysunThe present work is an attempt to design and implement an artificial game environment that provides infor-mation dissemination among Non-Player Characters (NPCs) of Open-World Role-Playing Game communities. Based on the perspective that an NPC is a mobile, communicating, cooperative agent that has autonomy and learning ability, and shows socialness in its virtual environment, it was possible to experiment and observe the impacts of human player-generated events on NPCs' social learning. By means of introducing different tuneable NPC autonomy and socialness parameter values, it is ensured that information and NPC opinions about the player not only spread over the society, but they also help develop player reputation among NPCs.The defined primitives such as NPCs' self and group-model views; event valuation and reputation models; established learning, memorization, forgetting mechanisms; the introduced information exchange and update protocol; and reputation metric provided for the construction of a tuneable, scalable virtual environment that can be used to investigate the individual and social behavioural aspects of artificial NPC societies. In this virtual environment, it is shown that player's reputation converges to fixed values. This is noteworthy since it is an indicator of societal learning and belief formation through NPC communications in the game environment.
