Fuzzy rule interpolation and reinforcement learning

@article{Vincze2017FuzzyRI,
  title={Fuzzy rule interpolation and reinforcement learning},
  author={D{\'a}vid Vincze},
  journal={2017 IEEE 15th International Symposium on Applied Machine Intelligence and Informatics (SAMI)},
  year={2017},
  pages={000173-000178}
}
  • D. Vincze
  • Published 2017
  • Computer Science
  • 2017 IEEE 15th International Symposium on Applied Machine Intelligence and Informatics (SAMI)
Reinforcement Learning (RL) methods became popular decades ago and still maintain to be one of the mainstream topics in computational intelligence. Countless different RL methods and variants can be found in the literature, each one having its own advantages and disadvantages in a specific application domain. Representation of the revealed knowledge can be realized in several ways depending on the exact RL method, including e.g. simple discrete Q-tables, fuzzy rule-bases, artificial neural… 

Tables from this paper

Antecedent Redundancy Exploitation in Fuzzy Rule Interpolation-based Reinforcement Learning

Novel methods which could improve the efficiency of the automated knowledge extraction methods used in the FRIQ-learning (Fuzzy Rule Interpolationbased Q-learning) machine learning method by facilitating the creation of a sparse fuzzy rule-base from which the knowledge can be directly extracted.

Parallelization by Vectorization in Fuzzy Rule Interpolation Adapted to FRIQ-Learning

  • D. Vincze
  • Computer Science
    2018 World Symposium on Digital Intelligence for Systems and Machines (DISA)
  • 2018
The goal was to identify parallelization possibilities and exploit fine-grained parallelism in FRIQ-learning by rearranging data structures and with using the Advanced Vector Extensions 2 (AVX2) SIMD instruction set present in most modern x86 architecture processors.

Applying Expert Heuristic as an a Priori Knowledge for FRIQ-Learning

The main goal of this paper is the introduction of a methodology, suitable for merging the a priori stateaction fuzzy control rule-base to the initial state-action-value function (Q-function) representation.

Expert heuristic tuning design for the FRIQ-learning

The main goal of this paper is to suggest a method for the FRIQ-learning system which may be suitable for optimizing the injected expert knowledgebase (Q-function) too, which is able to optimize (tune) the external knowledge rule-base during the learning phase too.

Clustering-based fuzzy knowledgebase reduction in the FRIQ-learning

  • T. TompaS. Kovács
  • Computer Science
    2017 IEEE 15th International Symposium on Applied Machine Intelligence and Informatics (SAMI)
  • 2017
A new, clustering based reduction method is introduced, suitable for eliminating the unnecessary rules of the rule-base and hence decrease the size of the fuzzy knowledgebase.

Tudásbázis redukció a szakértői szabályrendszerrel bővített FRIQ-learning módszerben

The main goal of this paper is to introduce a rule-base reduction strategy of the expert knowledgeincluded FRIQ-learning, which is able to reduce the rule- base size during the construction (learning) phase.

SZAKÉRTŐI HEURISZTIKA ALKALMAZÁSA A FRIQ-LEARNING MEGERŐSÍTÉSES TANULÁSI MÓDSZERBEN

The main goal of the paper is to introduce the new developed version of the FRIQ-learning, which starts the learning phase with not an empty knowledgebase but with an expert-defined, a priori knowledgebase.

Determining the minimally allowed rule-distance for the incremental rule-base contruction phase of the FRIQ-learning

  • T. TompaS. Kovács
  • Computer Science
    2018 19th International Carpathian Control Conference (ICCC)
  • 2018
A new rule-distance limit calculation methodology is introduced which can avoid the unnecessary high number of inserted rules in the initial phase of the iteration of the FRIQ-learning.

Football Simulation Modeling with Fuzzy Rule Interpolation-based Fuzzy Automaton

The goal of this work was to construct such a model, which employs a human-readable knowledge representation to control the agents in a football simulation, which can be adapted to real robot hardware and also can be used as a reference model for fuzzy logic based machine learning methods.

Smart Image-Processing based Energy Harvesting for Green Internet of Things

A FQL-based approach that can maximize the lifetime of sensors and accelerate the process of wireless energy harvesting (EH) for mobile sensors which coexist with macro and small base stations deployed over a time-variant heterogeneous network (HetNet).

References

SHOWING 1-10 OF 40 REFERENCES

Incremental Rule Base Creation with Fuzzy Rule Interpolation-Based Q-Learning

A method which can construct the requested FRI fuzzy model from scratch in a reduced size is introduced, achieved by incremental creation of an intentionally sparse fuzzy rule base.

Fuzzy Rule Interpolation-based Q-learning

  • D. VinczeS. Kovács
  • Computer Science
    2009 5th International Symposium on Applied Computational Intelligence and Informatics
  • 2009
The main goal of this paper is to introduce Fuzzy Rule Interpolation (FRI), namely the F Five (Fuzzy rule Interpolations based on Vague Environment) to be the model applied with Q-learning (F RIQ-learning).

Reduced Rule Base in Fuzzy Rule Interpolation-based Q-learning

The main goal of this paper is the introduction of a method which can construct the requested FRI fuzzy model in a reduced size by incremental creation of an intentionally sparse fuzzy rule base.

Experience-based rule base generation and adaptation for fuzzy interpolation

This paper presents a novel rule base generation and adaptation system to allow the creation of rule bases with minimal a priori knowledge by adding accurate interpolated rules into the rule base guided by a performance index from the feedback mechanism.

Extending the Fuzzy Rule Interpolation "FIVE" by Fuzzy Observation

The main contribution of this paper is the introduction of a way for handling fuzzy observations by extending the original “FIVE” concept with the ability of merging vague environments.

Rule-base reduction in Fuzzy Rule Interpolation-based Q-learning

The goal of the paper is to introduce possible methods, which aim to find and remove the redundant and unnecessary rules from the rule-base automatically by using variations of newly developed decremental rule base reduction strategies.

Reinforcement Learning: An Introduction

This book provides a clear and simple account of the key ideas and algorithms of reinforcement learning, which ranges from the history of the field's intellectual foundations to the most recent developments and applications.

Performance Optimization of the Fuzzy Rule Interpolation Method "FIVE"

The goal of this paper is to introduce some implementation details of a low-computation and lowresource-demand FRI method, together with its brief time and space complexity analysis.

A generalized concept for fuzzy rule interpolation

This paper proposes an interpolation methodology, whose key idea is based on the interpolation of relations instead of interpolating /spl alpha/-cut distances, and which offers a way to derive a family of interpolation methods capable of eliminating some typical deficiencies of fuzzy rule interpolation techniques.

Extracting symbolic knowledge from recurrent neural networks - A fuzzy logic approach