In order to minimize the generalization error in neural networks, a novel technique to identify overfitting phenomena when training the learner is formally introduced. This enables support of a reliable and trustworthy early stopping condition, thus improving the predictive power of that type of modeling. Our proposal exploits the correlation over time in a collection of online indicators, namely characteristic functions for indicating if a set of hypotheses are met, associated with a range of independent stopping conditions built from a canary judgment to evaluate the presence of overfitting. That way, we provide a formal basis for decision making in terms of interrupting the learning process. As opposed to previous approaches focused on a single criterion, we take advantage of subsidiarities between independent assessments, thus seeking both a wider operating range and greater diagnostic reliability. With a view to illustrating the effectiveness of the halting condition described, we choose to work in the sphere of natural language processing, an operational continuum increasingly based on machine learning. As a case study, we focus on parser generation, one of the most demanding and complex tasks in the domain. The selection of cross-validation as a canary function enables an actual comparison with the most representative early stopping conditions based on overfitting identification, pointing to a promising start toward an optimal bias and variance control.

We introduce an adaptive scheduling for adaptive sampling as a novel way of machine learning in the construction of part-of-speech taggers. The goal is to speed up the training on large data sets, without significant loss of performance with regard to an optimal configuration. In contrast to previous methods using a random, fixed or regularly rising spacing between the instances, ours analyzes the shape of the learning curve geometrically in conjunction with a functional model to increase or decrease it at any time. The algorithm proves to be formally correct regarding our working hypotheses. Namely, given a case, the following one is the nearest ensuring a net gain of learning ability from the former, it being possible to modulate the level of requirement for this condition. We also improve the robustness of sampling by paying greater attention to those regions of the training data base subject to a temporary inflation in performance, thus preventing the learning from stopping prematurely. The proposal has been evaluated on the basis of its reliability to identify the convergence of models, corroborating our expectations. While a concrete halting condition is used for testing, users can choose any condition whatsoever to suit their own specific needs.

Abstract We present a compiler that can be used to automatically obtain efficient Java implementations of parsing algorithms from formal specifications expressed as parsing schemata. The system performs an analysis of the inference rules in the input schemata in order to determine the best data structures and indexes to use, and to ensure that the generated implementations are efficient. The system described is general enough to be able to handle all kinds of schemata for different grammar formalisms, such as context‐free grammars and tree‐adjoining grammars, and it provides an extensibility mechanism allowing the user to define custom notational elements. This compiler has proven very useful for analyzing, prototyping and comparing natural‐language parsers in real domains, as can be seen in the empirical examples provided at the end of the paper. Copyright © 2008 John Wiley & Sons, Ltd.

Information Retrieval systems are limited by the linguistic variation of language. The use of Natural Language Processing techniques to manage this problem has been studied for a long time, but mainly focusing on English. In this paper we deal with European languages, taking Spanish as a case in point. Two different sources of syntactic information, queries and documents, are studied in order to increase the performance of Information Retrieval systems.