Evolution of Voronoi based Fuzzy Recurrent Controllers

Carlos Kavka,Patricia Roggero,Marc Schoenauer
DOI: https://doi.org/10.48550/arXiv.cs/0511091
2005-11-28
Abstract:A fuzzy controller is usually designed by formulating the knowledge of a human expert into a set of linguistic variables and fuzzy rules. Among the most successful methods to automate the fuzzy controllers development process are evolutionary algorithms. In this work, we propose the Recurrent Fuzzy Voronoi (RFV) model, a representation for recurrent fuzzy systems. It is an extension of the FV model proposed by Kavka and Schoenauer that extends the application domain to include temporal problems. The FV model is a representation for fuzzy controllers based on Voronoi diagrams that can represent fuzzy systems with synergistic rules, fulfilling the $\epsilon$-completeness property and providing a simple way to introduce a priory knowledge. In the proposed representation, the temporal relations are embedded by including internal units that provide feedback by connecting outputs to inputs. These internal units act as memory elements. In the RFV model, the semantic of the internal units can be specified together with the a priori rules. The geometric interpretation of the rules allows the use of geometric variational operators during the evolution. The representation and the algorithms are validated in two problems in the area of system identification and evolutionary robotics.
Artificial Intelligence
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is to explore the gain/loss asymmetry in the stock market and the reasons why this asymmetry exists in the Dow Jones Industrial Average (DJIA) but not in the individual stocks that make up the index. Specifically, by analyzing the investment time distribution (i.e., the distribution of the waiting time required to reach a predetermined return level), the authors found that for the entire DJIA, the time required to achieve the same percentage of positive or negative returns is different, indicating that there is some kind of synchronization effect or cascading process in the market as a whole, and this effect is not obvious when analyzing individual stocks separately. The author further explored the possible mechanisms behind this phenomenon, especially how psychological factors affect the overall behavior of the market and how this behavior leads to the differences between the market index and individual stocks. ### Summary of the core issues of the paper: 1. **Gain/Loss Asymmetry**: The study found that in the DJIA, the time required to achieve a positive return of the same magnitude is usually longer than the time required to achieve a negative return of the same magnitude, which is called gain/loss asymmetry. 2. **Source of Asymmetry**: Further research found that this asymmetry is significant in the DJIA but not obvious in the individual stocks that make up the DJIA, suggesting that this asymmetry may be caused by the overall behavioral characteristics of the market, such as the synchronization effect between different economic sectors. 3. **Influence of Psychological Factors**: It explored how human psychological factors, especially risk - averse behavior, work in the market, leading to the rapid decline and slow rise of the overall market. Through these analyses, the paper aims to provide a deeper understanding of the complex behaviors in the financial market, especially the differences between the overall market behavior and the behavior of individual stocks and the psychological and socio - economic mechanisms behind them.