Logo des Repositoriums

Auflistung nach Schlagwort "genetic algorithm"

1 - 3 von 3
  • Konferenzbeitrag
    Community Detection in Complex Networks using Genetic Algorithms
    (SKILL 2018 - Studierendenkonferenz Informatik, 2018) Lehnerer, Simon
    Detecting the community structure is of great interest when analyzing the topology of a network, however it is not a trivial problem. In this article a genetic algorithm is proposed which Ąnds the community structure of a network based on the maximization of a quality function called modularity. Tests using several sample networks show that it reliably Ąnds the community structure. However it does not resolve sufficiently small communities as intuitively expected due to an effect known as resolution limit.
  • Konferenzbeitrag
    Hyper-Parameter Search for Convolutional Neural Networks - An Evolutionary Approach
    (SKILL 2018 - Studierendenkonferenz Informatik, 2018) Bibaeva, Victoria
    Convolutional neural networks is one of the most popular neural network classes within the deep learning research area. Due to their specific architecture they are widely used to solve such challenging tasks as image and speech recognition, video analysis etc. The architecture itself is defined by a number of (hyper-)parameters that have major impact on the recognition rate. Although much significant progress has been made to improve the performance of convolutional networks, the typical hyper-parameter search is done manually, taking therefore a long time and likely to disregard some very good values. This paper solves the problem by proposing two different evolutionary algorithms for automated hyper-parameter search in convolutional architectures. It will be shown that in case of image recognition these algorithms are capable of finding architectures with nearly state of the art performance automatically, sparing the scientists from much tedious effort.
  • Konferenzbeitrag
    Towards Entropy-Based Requirements Elicitation
    (Information systems technology and its applications – 6th international conference – ISTA 2007, 2007) Kostanyan, Arkadiy V.; Shekhovtsov, Vladimir A.
    This paper presents our views towards constructing the universal parsing technique for the software requirements texts and the requirements elicitation technique based on the output of this parsing process. With the proposed parsing technique, it should be possible to achieve language-independent processing of the requirements texts. Source sentences are treated as systems with words as elements and with a state determining the set of links between elements. The main goal of the parser is to find the state of the system with the highest organization level by minimizing its entropy. The requirements elicitation technique extends trained indicator approach by Cleland-Huang treating inter-word links as indicators. The output of the elicitation process is the set of requirements information ready to be processed further using Conceptual and Aspectual Predesign techniques.