Search

This paper presents the first computational construction grammar for English that works for both comprehension and production – implemented in the open-source formalism Fluid Construction Grammar (Steels 2011, Beuls and van Trijp, 2016, www.fcg-net.org). This implementation is important because it demonstrates that the many salient characteristics of the constructional approach can be scaled to fully operational, broad-coverage grammars, which has been a matter of concern for mainstream linguists because of the unrestricted expressive power of constructions. As such, it offers a major contribution to the growing body of formal and computational work in the CxG community (e.g. Steels 2004, 2011, Bergen and Chang 2005, Boas and Sag 2012). The Basic ENglish Grammar (BENG) is conceived as a usage-based grammar in which constructions and their processing are intricately linked. The constructions themselves – partly hand-written and partly learned from data – range from highly schematic to highly specific constructions, which can be freely combined with each other as long as there are no conflicts (van Trijp 2016). The grammar includes a couple of dozen handwritten constructions for nominal and verbal phrases, for handling argument structure and information structure relations, and for handling basic speech acts. Other constructions are automatically extracted from lexical databases or acquired online through a set of learning operators. Language processing is implemented as a search process (Van Eecke and Beuls 2017) that tries to come up with the best possible verbalization of a given conceptualization (= production), or with the most plausible meaning for a given utterance (= comprehension). Crucially while processing, the FCG-system automatically builds constructional dependency graphs that show which constructions triggered the activation of which other constructions (see Figure 1). Besides demonstrating that the constructional perspective can be scaled, the English grammar provides crucial insights into the computational properties of constructions, and offers many potential applications such as information retrieval. Interested researchers can test the grammar online at http://www.fcg-net.org/fcg- interactive/.

International audience; Human languages have multiple strategies that allow us to discriminate objects in a vast variety of contexts. Colours have been extensively studied from this point of view. In particular, previous research in artificial language evolution has shown how artificial languages may emerge based on specific strategies to distinguish colours. Still, it has not been shown how several strategies of diverse complexity can be autonomously managed by artificial agents. We propose an intrinsic motivation system that allows agents in a population to create a shared artificial language and progressively increase its expressive power. Our results show that with such a system agents successfully regulate their language development, which indicates a relation between population size and consistency in the emergent communicative systems.

Trabajo presentado en la 10th International Conference on the Evolution of Language (EVOLANG 2014), celebrada en Viena del 14 al 17 de abril de 2014.

Steels, Luc (ed.)., This state-of-the-art-survey documents the Fluid Construction Grammar (FCG), a new formalism for the representation of lexicons and grammars, which has been used in a wide range of case studies for different languages, both for studying specific grammatical phenomena and design patterns, as for investigating language learning and language evolution. The book focuses on the many complex computational issues that arise when writing challenging real world grammars and hence emphasises depth of analysis rather than broad scope. The volume contains 13 contributions organized in 5 parts from "Basic", and "Implementation", over "Case Studies", and "Formal Analysis", up to 3 papers presenting a "Conclusion".

The fascinating question of the origins and evolution of language has been drawing a lot of attention recently, not only from linguists, but also from anthropologists, evolutionary biologists, and brain scientists. This groundbreaking book explores the cultural side of language evolution. It proposes a new overarching framework based on linguistic selection and self-organization and explores it in depth through sophisticated computer simulations and robotic experiments. Each case study investigates how a particular type of language system can emerge in a population of language game playing agents and how it can continue to evolve in order to cope with changes in ecological conditions. Case studies cover on the one hand the emergence of concepts and words for proper names, color terms, names for bodily actions, spatial terms and multi-dimensional words. The second set of experiments focuses on the emergence of grammar, specifically case grammar for expressing argument structure, functional grammar for expressing different uses of spatial relations, internal agreement systems for marking constituent structure, morphological expression of aspect, and quantifiers expressed as articles. The book is ideally suited as study material for an advanced course on language evolution and it will be of interest to anyone who wonders how human languages may have originated.

This document serves as a technical documentation of the Babel2 framework. It is work in progress and it is our aim to continuously improve and enhance the text. This manual accompanies Babel2 which can be downloaded for free at http://arti.vub.ac.be/b2dl/. Furthermore, http://fcg-net.org contains additional background information and downloadable papers that show in-depth experiments with Babel2. Babel2 connects the implementations of our core technologies such as Fluid Construction Grammar (FCG) and Incremental Recruitment Language (IRL) with mechanisms for multi-agent interactions, robotic embodiment, cognitive processing and learning. An extensive monitoring system gives access to every detail of Babel2's intermediate representations and dynamics and a high modularity ensures that the system can be used in a very wide variety of scenarios. Babel2 is written in Common Lisp and runs in all major Lisp implementations on all major plat- forms. Its source code is frequently released to the public under theGNU General Public License.