Byrski, Aleksander - ed. ; Kisiel-Dorohinicki, Marek - ed. ; Dobrowolski, Grzegorz - ed.
In this paper we propose a strategy learning model for autonomous agents based on classification. In the literature, the most commonly used learning method in agent-based systems is reinforcement learning. In our opinion, classification can be considered a good alternative. This type of supervised learning can be used to generate a classifier that allows the agent to choose an appropriate action for execution. Experimental results show that this model can be successfully applied for strategy generation even if rewards are delayed. ; We compare the efficiency of the proposed model and reinforcement learning using the farmer-pest domain and configurations of various complexity. In complex environments, supervised learning can improve the performance of agents much faster that reinforcement learning. If an appropriate knowledge representation is used, the learned knowledge may be analyzed by humans, which allows tracking the learning process.
Zielona Góra: Uniwersytet Zielonogórski
AMCS, volume 25, number 3 (2015) ; kliknij tutaj, żeby przejść
Biblioteka Uniwersytetu Zielonogórskiego
2024-05-02
2024-05-02
55
https://zbc.uz.zgora.pl/repozytorium/publication/88894
Nazwa wydania | Data |
---|---|
A strategy learning model for autonomous agents based on classification | 2024-05-02 |
Kajdanowicz, Tomasz Kazienko, Przemysław Cordón, Oskar - ed. Kazienko, Przemysław - ed.
Ponulak, Filip Kasiński, Andrzej - ed. Ponulak, Filip - ed.
Zajdel, Roman Korbicz, Józef (1951- ) - red. Uciński, Dariusz - red.
Kasiński, Andrzej Ponulak, Filip Korbicz, Józef (1951- ) - red.
Biedrzycki, Rafał Arabas, Jarosław Korbicz, Józef (1951- ) - red. Uciński, Dariusz - red.
Doya, Kenji Kimura, Hidenori Miyamura, Aiko Fliess, Michel - ed. Jai, Abdelhaq El - ed.
Czekalski, Piotr Korbicz, Józef (1951- ) - red. Uciński, Dariusz - red.
Świniarski, Roman W. Grzymala-Busse, Jerzy - ed. Świniarski, Roman W. - ed. Zhong, Ning - ed. Ziarko, Wojciech - ed.