An Integrated Fuzzy Logic System in A Partially Known Environment
Cited 0 times inCited 0 times in
- An Integrated Fuzzy Logic System in A Partially Known Environment
- Shin, Kyuyeol
- Bien, Zeungnam
- Issue Date
- Graduate School of UNIST
- In this thesis, we deal with a learning method for mimicking human behaviors in a partially known environment. Humans are possible to infer the appropriate action based on the partially known information to achieve their goals. The fuzzy logic is able to deal with a process of human reasoning by describing IF-THEN rules. The FQL method which is one of the reinforcement learning methods uses fuzzy logic for decision making. FQL is possible to expect improvement of the learning efficiency rather than ordinary Q-Learning method by using fuzzy logic. However, the problem comes from a conclusion part of FQL. The conclusion part of FQL consists of a set of singleton values. The behaviors of the agent are influenced by the number of singleton values. Especially, it is hart to expect the efficiency of learning when the environment is complex and a precise behavior is required. So, we suggest an integrated fuzzy logic system which is FQL with ANFIS.
- Electrical Engineering
- ; Go to Link
- Appears in Collections:
- Files in This Item:
An Integrated Fuzzy Logic System in A Partially Known Environment.pdf
can give you direct access to the published full text of this article. (UNISTARs only)
Show full item record
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.