• Journal of KIISE:Software and Applications
• /
• v.31 no.11
• /
• pp.1431-1438
• /
• 2004

Reinforcement learning learns policies for accomplishing a task's goal by experience through interaction between agent and environment. Q-learning, basis algorithm of reinforcement learning, has the problem of curse of dimensionality and slow learning speed in the incipient stage of learning. In order to solve the problems of Q-learning, new function approximation methods suitable for reinforcement learning should be studied. In this paper, to improve these problems, we suggest Fuzzy Q-Map algorithm that is based on online fuzzy clustering. Fuzzy Q-Map is a function approximation method suitable to reinforcement learning that can do on-line teaming and express uncertainty of environment. We made an experiment on the mountain car problem with fuzzy Q-Map, and its results show that learning speed is accelerated in the incipient stage of learning.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.6
• /
• pp.635-642
• /
• 2003

Reinforcement learning got successful results in a lot of applications such as control and scheduling. Various function approximation methods have been studied in order to improve the learning speed and to solve the shortage of storage in the standard reinforcement learning algorithm of Q-Learning. Most function approximation methods remove some special quality of reinforcement learning and need prior knowledge and preprocessing. Fuzzy Q-Learning needs preprocessing to define fuzzy variables and Local Weighted Regression uses training examples. In this paper, we propose a function approximation method, Fuzzy Q-Map that is based on on-line fuzzy clustering. Fuzzy Q-Map classifies a query state and predicts a suitable action according to the membership degree. We applied the Fuzzy Q-Map, CMAC and LWR to the mountain car problem. Fuzzy Q-Map reached the optimal prediction rate faster than CMAC and the lower prediction rate was seen than LWR that uses training example.

• Transactions on Control, Automation and Systems Engineering
• /
• v.3 no.3
• /
• pp.170-175
• /
• 2001

Q-learning is a kind of reinforcement learning where the agent solves the given task based on rewards received from the environment. Most research done in the field of Q-learning has focused on discrete domains, although the environment with which the agent must interact is generally continuous. Thus we need to devise some methods that enable Q-learning to be applicable to the continuous problem domain. In this paper, an extended fuzzy rule is proposed so that it can incorporate Q-learning. The interpolation technique, which is widely used in memory-based learning, is adopted to represent the appropriate Q value for current state and action pair in each extended fuzzy rule. The resulting structure based on the fuzzy inference system has the capability of solving the continuous state about the environment. The effectiveness of the proposed structure is shown through simulation on the cart-pole system.

• The KIPS Transactions:PartB
• /
• v.10B no.6
• /
• pp.587-592
• /
• 2003

Many real world control problems have continuous states and actions. When the state space is continuous, the reinforcement learning problems involve very large state space and suffer from memory and time for learning all individual state-action values. These problems need function approximators that reason action about new state from previously experienced states. We introduce Fuzzy Q-Map that is a function approximators for 1 - step Q-learning and is based on fuzzy clustering. Fuzzy Q-Map groups similar states and chooses an action and refers Q value according to membership degree. The centroid and Q value of winner cluster is updated using membership degree and TD(Temporal Difference) error. We applied Fuzzy Q-Map to the mountain car problem and acquired accelerated learning speed.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04b
• /
• pp.271-273
• /
• 2000

일반적으로, 지능형 에이전트는 사용자의 목적과 주위 환경으로부터 최적의 행동을 스스로 찾아낼 수 있어야 한다. 만약 에이전트의 목적이나 주위 환경이 불확실성을 포함하는 경우, 에이전트는 적절한 행동을 선택하기 어렵다. 그러나, 사용자의 목적이 인간 지식의 불확실성을 포함하는 언어값으로 표현되었을 경우, 이를 처리하려는 연구는 없었다. 본 논문에서는 모호한 사용자의 의도를 퍼지 목적으로 나타내고, 에이전트가 인지하는 불확실한 환경을 퍼지 상태로 표현하는 방법을 제안한다. 또, 퍼지 목적과 상태를 이용하여 확장한 펴지 강화 함수와를 이용하여, 기존 강화 학습 알고리즘 중 하나인 Q-Learning을 FuzzyQ-Learning으로 확장하고, 이에 대한 타당성을 검증한다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.163-167
• /
• 2000

The eligibility is used to solve the credit-assignment problem which is one of important problems in reinforcement learning. Conventional eligibilities which are accumulating eligibility and replacing eligibility make ineffective use of rewards acquired in learning process. Because only an executed action in a visited state is learned by these eligibilities. Thus, we propose a new eligibility, called the weighted eligibility with which not only an executed action but also neighboring actions in a visited state are to be learned. The fuzzy Q-learning algorithm using proposed eligibility is applied to a cart-pole balancing problem, which shows improvement of learning speed.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.5
• /
• pp.388-394
• /
• 2001

Reinforcement learning is a kind of unsupervised learning methods that an agent control rules from experiences acquired by interactions with environment. The eligibility is used to resolve the credit-assignment problem which is one of important problems in reinforcement learning, Conventional eligibilities such as the accumulating eligibility and the replacing eligibility are ineffective in use of rewards acquired in learning process, since on1y one executed action for a visited state is learned. In this paper, we propose a new eligibility, called the distributed eligibility, with which not only an executed action but also neighboring actions in a visited state are to be learned. The fuzzy Q-learning algorithm using the proposed eligibility is applied to a cart-pole balancing problem, which shows the superiority of the proposed method to conventional methods in terms of learning speed.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1300-1304
• /
• 2003

The following paper deals with a control problem of a mobile robot in home network environment. The home network causes the mobile robot to communicate with sensors to get the sensor measurements and to be adapted to the environment changes. To get the improved performance of control of a mobile robot in spite of the change in home network environment, we use the fuzzy inference system with multiple reward reinforcement learning. The multiple reward reinforcement learning enables the mobile robot to consider the multiple control objectives and adapt itself to the change in home network environment. Multiple reward fuzzy Q-learning method is proposed for the multiple reward reinforcement learning. Multiple Q-values are considered and max-min optimization is applied to get the improved fuzzy rule. To show the effectiveness of the proposed method, some simulation results are given, which are performed in home network environment, i.e., LAN, wireless LAN, etc.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.8 no.3
• /
• pp.220-224
• /
• 2008

This paper presents the polygon-based Q-leaning and Parallel SVM algorithm for object search with multiple robots. We organized an experimental environment with one hundred mobile robots, two hundred obstacles, and ten objects. Then we sent the robots to a hallway, where some obstacles were lying about, to search for a hidden object. In experiment, we used four different control methods: a random search, a fusion model with Distance-based action making (DBAM) and Area-based action making (ABAM) process to determine the next action of the robots, and hexagon-based Q-learning, and dodecagon-based Q-learning and parallel SVM algorithm to enhance the fusion model with Distance-based action making (DBAM) and Area-based action making (ABAM) process. In this paper, the result show that dodecagon-based Q-learning and parallel SVM algorithm is better than the other algorithm to tracking for object.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.1358-1361
• /
• 1993