This article considers the problem of learning the correct temporal sequence of discrete behaviors from a finite behavior set that will lead to completion of a complex task, using only stochastic reinforcement from the environment. A trial-and-error learning algorithm is proposed that is inspired by backward chaining from the animal training discipline. The procedure is analytically formulated ...

In this paper we consider uncountable classes recognizable by ω-automata and investigate suitable learning paradigms for them. In particular, the counterparts of explanatory, vacillatory and behaviourally correct learning are introduced for this setting. Here the learner reads in parallel the data of a text for a language L from the class plus an ω-index α and outputs a sequence of ω-automata s...

Automata learning enables model-based analysis of black-box systems by automatically constructing models from system observations, which are often collected via testing. The required testing budget to learn adequate heavily depends on the applied and techniques. Test cases executed for (1) collect behavioural information (2) falsify learned hypothesis automata. Falsification test-cases commonly...

This paper aims to introduce an effective classification method of learning for partitioning the data in statistical spaces. The work is based on using multi-constraint partitioning on the stochastic learning automata. Stochastic learning automata with fixed or variable structures are a reinforcement learning method. Having no information about optimized operation, such models try to find an an...

The area of automata learning was pioneered by Angluin in the 80’s [1]. Her original algorithm, which applied to regular languages and deterministic automata, has been extended to various types of automata and used in software and hardware verification. In this talk, we will take an abstract perspective at automata learning. We show how the correctness of the original algorithm and many extensi...

We present a new algorithm IDS for incremental learning of deterministic finite automata (DFA). This algorithm is based on the concept of distinguishing sequences introduced in [1]. We give a rigorous proof that two versions of this learning algorithm correctly learn in the limit. Finally we present an empirical performance analysis that compares these two algorithms, focussing on learning time...

Learning automata are used at the source nodes of a multimedia network to dynamically control the externally arriving traffic to the network. A new fixed structure learning automaton algorithm is used to control the traffic. At every network source, each different traffic type (data, voice, video, e.t.c.) is controlled by a different learning automaton. Every learning automaton makes its decisi...

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 har...