On Some Asymptotic Properties of Learning Automaton Networks
نویسنده
چکیده
In this report, we analyze the collective behavior of learning automata which are used in a programming language under development that combines reinforcement learning and symbolic programming [2, 6]. Learning automata can automatically improve their behavior by using a response from a random stationary environment, but when connected with each other, their behavior becomes much complex and hard to analyze. We analyzed a class of learning automaton networks and proved that they eventually take the most rewarding action with probability one when they use an appropriately decaying learning rate.
منابع مشابه
An automaton group: a computational case study
We introduce a two generated weakly branch contracting automaton group $G$ which is generated by a two state automaton on a three letter alphabet. Using its branch structure and the finiteness nature of a sequence of its factor groups we compute the order of some of these factors. Furthermore some algebraic properties of $G$ are detected .
متن کاملDeterministic Fuzzy Automaton on Subclasses of Fuzzy Regular ω-Languages
In formal language theory, we are mainly interested in the natural language computational aspects of ω-languages. Therefore in this respect it is convenient to consider fuzzy ω-languages. In this paper, we introduce two subclasses of fuzzy regular ω-languages called fuzzy n-local ω-languages and Buchi fuzzy n-local ω-languages, and give some closure properties for those subclasses. We define a ...
متن کاملBifurcations of Recurrent Neural Networks in Gradient Descent Learning
Asymptotic behavior of a recurrent neural network changes qualitatively at certain points in the parameter space, which are known as \bifurcation points". At bifurcation points, the output of a network can change discontinuously with the change of parameters and therefore convergence of gradient descent algorithms is not guaranteed. Furthermore, learning equations used for error gradient estima...
متن کاملAsymptotic Close to Optimal Resource Allocation in Centralized Multi-band Wireless Networks
This paper concerns sub-channel allocation in multi-user wireless networks with a view to increasing the network throughput. It is assumed there are some sub-channels to be equally divided among active links, such that the total sum rate increases, where it is assumed each link is subject to a maximum transmit power constraint. This problem is found to be a non-convex optimization problem and i...
متن کاملUsing an Evaluator Fixed Structure Learning Automata in Sampling of Social Networks
Social networks are streaming, diverse and include a wide range of edges so that continuously evolves over time and formed by the activities among users (such as tweets, emails, etc.), where each activity among its users, adds an edge to the network graph. Despite their popularities, the dynamicity and large size of most social networks make it difficult or impossible to study the entire networ...
متن کامل