Jouffe, L., Fuzzy Inference System Learning by Reinforcement Methods, IEEE Trans. on SMC-Part B, Vol. 28, No. 3, August 1998, pages 338-355. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Hybrid Computational Intelligence Schemes in Complex.. - Tsakonas, Dounias (2002) (Correct)
....knowledge acquisition is presented by [99] with the construction of Fuzzy ID3, an inductive decision tree generator. A very interesting study on heuristic algorithms for generating fuzzy decision trees is also presented by [100] Fuzzy sets and machine learning are also working together in [101], where a fuzzy inference system learning by reinforcement methods is presented. Finally, another similar attempt for combining fuzzy set theory and inductive machine learning is given in [102] 2.5 Machine Learning and Evolutionary algorithms Machine learning often works as feature selection or ....
Jouffe L.: Fuzzy Inference System Learning by Reinforcement Methods, IEEE Trans. on Systems Man & Cybernetics, PART C: Applications & Reviews, Vol. 28, No. 3, Aug. 1998, pp. 338-355.
Fuzzy Q-Learning - Glorennec, Jouffe (1997) (1 citation) Self-citation (Jouffe) (Correct)
....ffl Prior knowledge can be embedded into the fuzzy rules, which can reduce training significantly. This paper presents one extension of Watkin s QLearning into a fuzzy environment. A more complete description of Fuzzy Q Learning (FQL) and Fuzzy Actor Critic Learning (FACL) can be found in [14]. This paper is organized as follows: in Section 2, we present the FQL algorithm. We compare FQL and a Genetic Algorithm approach in Section 3 with the Cart Centering Problem, and we conclude. 2 Fuzzy Q Learning 2.1 Presentation We have proposed earlier [12] to represent both the actions and ....
.... explore the set of possible actions and acquire experience through the reinforcement signals, the actions are selected using an Exploration Exploitation Policy (EEP) Instead of the usual Boltzmann exploration, we use either a Pseudo Stochastic [8] or a combined directed nondirected exploration [14], in order to maximize the escounted knowledge gain. Let i y be the selected action in rule i using an EEP and i such as q[i; i ] max jJ q[i; j] The actual Qvalue of the infered action, a, is: Q(x; a) P N i=1 ff i (x) Theta q[i; i y ] P N i=1 ff i (x) 5) and the value of ....
[Article contains additional citation context not shown here]
Jouffe L., "Fuzzy Inference System learning by reinforcement methods", Tech. Rep. INSA 96081, submitted, 1996.
Fuzzy Inference System Learning by Reinforcement Methods - Jouffe (1997) (2 citations) Self-citation (Jouffe) (Correct)
....; D(x; y) x; y) 2 Z v ; Gamma1 (x; y) 2 Z f ; 0 otherwise; where D is a function which gives a reinforcement decreasing linearly from 1 to Gamma1 relative to the distance from the success zone. A. 3 Results All the results presented here have been obtained after a statistical study done in [52]. In this study, an experiment with a set of parameter values consists in an average of the performance measures over 30 runs. A run is made of a learning phase and a testing phase . The learning phase ends when 40 successive non failure zones have been reached or when the number of trials ....
....of trajectory obtained with a fuzzy controller tuned by FACL or FQL. This corresponds to a testing phase in which there are no failure trials, although we see here only four percent of the testing departure points. Fig. 7. Trajectories obtained with a Fuzzy Controller tuned with these methods In [52], 576 experiments were used for the statistical analysis of FACL parameters, i.e. 17280 learning and testing phases. These experiments came from the use of different fixed learning rates, adaptive learning rates, recency factors (actor and critic) and exploration characteristics (see [52] for ....
[Article contains additional citation context not shown here]
L. Jouffe, Fuzzy Inference Systems Learning by Reinforcement Methods: Application to a Pig House Atmosphere Control, Ph.D. thesis, University of Rennes I, 1997, (in French).
Accuracy Based Fuzzy Q-Learning for Robot - Behaviours Dongbing Gu (Correct)
No context found.
Jouffe, L., Fuzzy Inference System Learning by Reinforcement Methods, IEEE Trans. on SMC-Part B, Vol. 28, No. 3, August 1998, pages 338-355.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC