A non parametric model in the construction of biologi cal networks



A macroscopic model for self-organized aggregation in swarm robotic systems
Soysal, Onur; Şahin, Erol (2006-10-01)
We study the self-organized aggregation of a swarm of robots in a closed arena. We assume that the perceptual range of the robots are smaller than the size of the arena and the robots do not have information on the size of the swarm or the arena. Using a probabilistic aggregation behavior model inspired from studies of social insects, we propose a macroscopic model for predicting the final distribution of aggregates in terms of the parameters of the aggregation behavior, the arena size and the sensing chara...
A Probabilistic geometric model of self-organized aggregation in swarm robotic systems
Bayındır, Levent; Şahin, Erol; Department of Computer Engineering (2012)
Self-organized aggregation is the global level gathering of randomly placed robots using local sensing. Developing high performance and scalable aggregation behaviors for a swarm of mobile robots is non-trivial and still in need, when robots control themselves, perceive only a small part of the arena, and do not have access to information such as their position, the size of the arena or the number of robots. In this thesis, we developed a non-spatial probabilistic geometric model for self-organized aggregat...
A computational model of memory processes in the expectation-violation effect
Özyörük, Nilüfer; Say, Bilge; Department of Cognitive Sciences (2004)
This thesis focuses on modeling Expectation-Violation Effect, which is the superior recall of weakly associated pairs of words over strongly associated pairs. The goal of this thesis is to provide an exploratory computational model. A virtual experiment is conducted based on the datasets used in the psychological experiment by Amster et al. (1992). The computational modeling of this phenomenon is carried in the medium of ACT-R cognitive architecture.
A longest path problem for evaluating the worst case packet delay of switched ethernet
Schmidt, Klaus Verner; Schmidt, Şenan Ece (2010-07-09)
In the recent years, the use of real-time Ethernet protocols becomes more and more relevant for time-critical networked industrial applications. In this context, this paper presents a method to compute the worst-case packet delays on switched Ethernet. Based on an evaluation of the packet delays at each switch port and the network topology, we construct a weighted directed graph that allows to find the worst-case end-to-end packet delay by solving a conventional longest-path problem.
A Theoretical Analysis of Feature Fusion in Stacked Generalization
Ozay, Mete; Yarman Vural, Fatoş Tunay (2009-04-11)
In the present work, a theoretical framework in order to define the general performance of stacked generalization learning algorithm is developed. Analytical relationships between the performance of the Stacked Generalization classifier relative to the individual classifiers are constructed by the proposed theorems and the practical techniques are developed in order to optimize the performance of stacked generalization algorithm based on these relationships.
Citation Formats
M. Ağraz and V. Purutçuoğlu Gazi, “A non parametric model in the construction of biologi cal networks,” 2016, Accessed: 00, 2021. [Online]. Available: https://hdl.handle.net/11511/76033.