Among humans, teaching various tasks is a complex process which relies on multiple means for interaction and learning, both on the part of the teacher and of the learner. Used together, these modalities lead to effective teaching and learning approaches, respectively. In the robotics domain, task teaching has been mostly addressed by using only one or very few of these interactions. In this paper we present an approach for teaching robots that relies on the key features and the general approach people use when teaching each other: first give a demonstration, then allow the learner to refine the acquired capabilities by practicing under the teacher's supervision, involving a small number of trials. Depending on the quality of the learned task, the teacher may either demonstrate it again or provide specific feedback during the learner's practice trial for further refinement. Also, as people do during demonstrations, the teacher can provide simple instructions and informative cues, increasing the performance of learning. Thus, instructive demonstrations, generalization over multiple demonstrations and practice trials are essential features for a successful human-robot teaching approach. We implemented a system that enables all these capabilities and validated these concepts with a Pioneer 2DX mobile robot learning tasks from multiple demonstrations and teacher feedback.

This paper revisits the theory of regular inference, in particular by extending the definition of structural completeness of a positive sample and by demonstrating two basic theorems. This framework enables to state the regular inference problem as a search through a boolean lattice built from the positive sample. Several properties of the search space are studied and generalization criteria are discussed. In this framework, the concept of border set is introduced, that is the set of the most general solutions excluding a negative sample. Finally, the complexity of regular language identification from both a theoritical and a practical point of view is discussed. 1 Introduction Regular inference is the process of learning a regular language from a set of examples, consisting of a positive sample, i.e. a finite subset of a regular language. A negative sample, i.e. a finite set of strings not belonging to this language, may also be available. This problem has been studied as early as th...

Determinism plays an important role in grammatical inference.

The World Wide Web is a treasure trove of information. The Web's sheer scale makes automatic location and extraction of information appealing. However, much of the information lies buried in documents designed for human consumption, such as home pages or product catalogs. Before software agents can extract nuggets of information from Web documents, they have to be able to recognize it despite the multitude of formats in which it may appear. In this paper, we take a machine learning approach to the problem. We explain why existing grammar inference techniques face difficulties in this domain, present a new technique, and demonstrate its success on examples drawn from the Web ranging from CMU Tech Report codes to bus schedules. Our algorithm is shown to learn target languages found on the Web in significantly fewer examples than previous methods. In addition, our algorithm is guaranteed to learn in the limit, and runs in time O(|S| 4 ), where |S| is the size of the sample. Introduction...

Inferring an appropriate DTD or XML Schema Definition (XSD) for a given collection of XML documents essentially reduces to learning deterministic regular expressions from sets of positive example words. Unfortunately, there is no algorithm capable of learning the complete class of deterministic regular expressions from positive examples only, as we will show. The regular expressions occurring in practical DTDs and XSDs, however, are such that every alphabet symbol occurs only a small number of times. As such, in practice it suffices to learn the subclass of regular expressions in which each alphabet symbol occurs at most k times, for some small k. We refer to such expressions as k-occurrence regular expressions (k-OREs for short). Motivated by this observation, we provide a probabilistic algorithm that learns k-OREs for increasing values of k, and selects the one that best describes the sample based on a Minimum Description Length argument. The effectiveness of the method is empirically validated both on real world and synthetic data. Furthermore, the method is shown to be conservative over the simpler classes of expressions considered in previous work.

Abstract We explore the formal foundations of recent studies comparing aural pattern recognition capabilities of populations of human and non-human animals. To date, these experiments have focused on the boundary between the Regular and Context-Free stringsets. We argue that experiments directed at distinguishing capabilities with respect to the Subregular Hierarchy, which subdivides the class of Regular stringsets, are likely to provide better evidence about the distinctions between the cognitive mechanisms of humans and those of other species. Moreover, the classes of the Subregular Hierarchy have the advantage of fully abstract descriptive (model-theoretic) characterizations in addition to characterizations in more familiar grammar- and automata-theoretic terms. Because the descriptive characterizations make no assumptions about implementation, they provide a sound basis for drawing conclusions about potential cognitive mechanisms from the experimental results. We review the Subregular Hierarchy and provide a concrete set of principles for the design and interpretation of these experiments. Keywords Sub-regular languages · Local languages · Artificial grammar learning · Cognitive complexity · Aural pattern recognition · Mathematics of language J. Rogers (B)

A genetic approach to the inference of stochastic regular grammars from a given finite set of sample words is presented. The goal of the inference problem is not only to find a grammar that covers the given finite sample, but possibly also the infinite language from which the sample was taken (generalization). We propose two different bitstring representation methods for stochastic regular grammars and have a closer look at the objective function. Due to the large complexity of the problem, a massively parallel implementation of genetic algorithms was used. The algorithm was applied to a workload-modelingproblem. The results are compared with reference methods like the successor-, k-tail- and k-TLSS-method. 1 INTRODUCTION Genetic algorithms have successfully been used as a powerful global optimization method for problems with a large search space and a multimodal or otherwise difficult objective function. One such problem is the construction of a grammar for a (finite or infinite) lan...

A k-Testable tree set in the Strict sense (k-TS) is essentially defined by a finite set of patterns of "size " k that are permitted to appear in the trees of the tree language. Given a positive sample S of trees over a ranked alphabet, an algorithm is proposed which obtains the smallest k-TS tree set containing S. The proposed algorithm is polynomial on the size of S and identifies the class of k-TS tree languages in the limit from positive data. I.

The structural limitations of N-Gram models used for Language Modelling are illustrated through several examples. In most cases of interest, these limitations can be easily overcome using (general) regular or finite-state models, without having to resort to more complex, recursive devices. The problem is how to obtain the required finite-state structures from reasonably small amounts of training (positive) sentences of the considered task. Here this problem is approached through a Grammatical Inference technique known as MGGI. This allows us to easily apply a priory knowledge about the type of syntactic constraints that are relevant to the considered task to significantly improve the performance of N-Grams, using similar or smaller amounts of training data. Speech Recognition experiments are presented with results supporting the interest of the proposed approach.

Probabilistic finite-state machines are used today in a variety of areas in pattern recognition, or in fields to which pattern recognition is linked: computational linguistics, machine learning, time series analysis, circuit testing, computational biology, speech recognition and machine translation are some of them. In part I of this paper we survey these generative objects and study their definitions and properties. In part II, we will study the relation of probabilistic finite-state automata with other well known devices that generate strings as hidden Markov models and n-grams, and provide theorems, algorithms and properties that represent a current state of the art of these objects.