China
Africa
Explore chapters and articles related to this topic
Stochastic Learning Automata (SLA)
Published in Phil Mars, J.R. Chen, Raghu Nambiar, Learning Algorithms, 1996
Phil Mars, J.R. Chen, Raghu Nambiar
The concept of Stochastic Automata was first introduced by the pioneering work of Tsetlin in the early 1960s in the Soviet Union. Tsetlin was interested in the modeling of the behaviour of biological systems [Tse62]. Subsequent research has considered the use of the learning paradigms in engineering systems. This has led to extensive work using automata as models of learning with applications in telephone routing, pattern recognition, object partitioning and adaptive control [NT74, Lak81, NT89a, OM88, SN69, FM66]. A Learning Automaton can be regarded as an abstract object having a finite number of actions. It operates by selecting an action from a finite set of actions which is then evaluated by a random environment. The response from the environment is used by the automaton to select the next action. By this process, the automaton learns asymptotically to select the optimal action. The manner in which the automaton uses the response from the environment to select its next action is determined by the specific learning algorithm used. The next section gives details of the components of a SLA.
QoS-Based Hierarchical Routing in ATM Networks Using Reinforcement Learning Algorithms: A Methodology
Published in Witold Pedrycz, Athanasios Vasilakos, Computational Intelligence in Telecommunications Networks, 2018
Marios P. Saltouros, Antonios F. Atlasis, Athanasios V. Vasilakos, Witold Pedrycz
Generally, a Learning Automaton (LA) is a finite-state machine that interacts with a stochastic environment, trying to learn the optimal action the environment offers through a learning process (Figure 10.2). At any iteration the automaton chooses an action, according to a probability vector, using an output function. This function stimulates the environment which responds with an answer (reward or penalty). The automaton takes into account this answer and jumps, if necessary, to a new state using a transition function.
Path planning for USV based on artificial potential field approach with learning automata
Published in Selma Ergin, C. Guedes Soares, Sustainable Development and Innovations in Marine Technologies, 2022
Chunshui Xiong, Wei Wu, Rui Wu, Dongming Zhao
Learning Automata (LA) is an intelligent unit based on augmented learning and accomplishes adaptive decision making in an unknown stochastic environment (Varricchio 2016, Narendra & Thathachar 1989). Through a finite number of iterations, the LA gradually learns the best action (Rezvanian, Moradabadi, Ghavipour, Khomami, & Meybodi 2019, Santharam, Sastry, & Thathachar 1994).
A Two-Level Function Evaluation Management Model for Multi-Population Methods in Dynamic Environments: Hierarchical Learning Automata Approach
Published in Journal of Experimental & Theoretical Artificial Intelligence, 2021
Javidan Kazemi Kordestani, Mohammad Reza Meybodi, Amir Masoud Rahmani
A learning automaton (Narendra & Thathachar, 1974, 2012) is an adaptive decision-making unit that improves its performance by learning the way to choose the optimal action from a finite set of allowable actions through iterative interactions with an unknown random environment. The action is chosen at random based on a probability distribution kept over the action-set and at each instant the given action is served as the input to the random environment. The environment responds the taken action in turn with a reinforcement signal. The action probability vector is updated based on the reinforcement feedback from the environment. The objective of a learning automaton is to find the optimal action from the action-set so that the average penalty received from the environment is minimised. Figure 1 shows the relationship between the learning automaton and random environment.