While neuroscientists garner success in identifying brain regions and in analyzing individual neurons, ground is still being broken at the intermediate scale of understanding how neurons combine to encode information. This book proposes a method of representing information in a computer that would be suited for modeling the brain's methods of processing information. Holographic Reduced Representations (HRRs) are introduced here to model how the brain distributes each piece of information among thousands of neurons. It had been previously thought that the grammatical structure of a language cannot be encoded practically in a distributed representation, but HRRs can overcome the problems of earlier proposals. Thus this work has implications for psychology, neuroscience, linguistics, and computer science, and engineering.
Current representation schemes for automatic text classification treat documents as syntactically unstructured collections of words (Bag-of-Words) or 'concepts' (Bag-of-Concepts). Past attempts to encode syntactic structure have treated part-of-speech information as another word-like feature, but have been shown to be less effective than non-structural approaches. We propose a new representation scheme using Holographic Reduced Representations (HRRs) as a technique to encode both semantic and syntactic structure, though in very different ways. This method is unique in the literature in that it encodes the structure across all features of the document vector while preserving text semantics. Our method does not increase the dimensionality of the document vectors, allowing for efficient computation and storage. We present the results of various Support Vector Machine classification experiments that demonstrate the superiority of this method over Bag-of-Concepts representations and improvement over Bag-of-Words in certain classification contexts.
This volume constitutes seleted papers from the 12th International Conference on Text, Speech and Dialogue, TSD 2009, held in Pilsen, Czech Republic, in September 2009. This volume contains a collection of submitted papers presented at the conference which were thoroughly reviewed by three members of the conference reviewing team consisting of more than 40 top specialists in the conference topic areas. A total of 53 accepted papers out of 112 submitted, altogether contributed 127 authors and co-authors, were selected for presentation at the conference by the program committee and then included in this book. Theoretical and more general contributions were presented in common (plenary) sessions. Problem oriented sessions as well as panel discussions then brought together the specialists in limited problem areas with the aim of exchanging knowledge and skills resulting from research projects of all kinds.
This book constitutes the refereed proceedings of the 7th International Conference on Rough Sets and Current Trends in Computing, RSCTC 2010, held in Warsaw, Poland, in June 2010.
Acquire the tools for understanding new architectures and algorithms of dynamical recurrent networks (DRNs) from this valuable field guide, which documents recent forays into artificial intelligence, control theory, and connectionism. This unbiased introduction to DRNs and their application to time-series problems (such as classification and prediction) provides a comprehensive overview of the recent explosion of leading research in this prolific field. A Field Guide to Dynamical Recurrent Networks emphasizes the issues driving the development of this class of network structures. It provides a solid foundation in DRN systems theory and practice using consistent notation and terminology. Theoretical presentations are supplemented with applications ranging from cognitive modeling to financial forecasting. A Field Guide to Dynamical Recurrent Networks will enable engineers, research scientists, academics, and graduate students to apply DRNs to various real-world problems and learn about different areas of active research. It provides both state-of-the-art information and a road map to the future of cutting-edge dynamical recurrent networks.
This volume features the complete text of the material presented at the Nineteenth Annual Conference of the Cognitive Science Society. Papers have been loosely grouped by topic and an author index is provided in the back. As in previous years, the symposium included an interesting mixture of papers on many topics from researchers with diverse backgrounds and different goals, presenting a multifaceted view of cognitive science. In hopes of facilitating searches of this work, an electronic index on the Internet's World Wide Web is provided. Titles, authors, and summaries of all the papers published here have been placed in an online database which may be freely searched by anyone. You can reach the web site at: www-csli.stanford.edu/cogsci97.
This book constitutes the refereed proceedings of the 8th International Conference on Flexible Query Answering Systems, FQAS 2009, held in Roskilde, Denmark, in October 2009. The 57 papers included in this volume were carefully reviewed and selected from 90 submissions. They are structured in topical sections on database management, information retrieval, extraction and mining, ontologies and semantic web, intelligent information extraction from texts, advances in fuzzy querying, personalization, preferences, context and recommendation, and Web as a stream.
Adaptivity and learning have in recent decades become a common concern of scientific disciplines. These issues have arisen in mathematics, physics, biology, informatics, economics, and other fields more or less simultaneously. The aim of this publication is the interdisciplinary discourse on the phenomenon of learning and adaptivity. Different perspectives are presented and compared to find fruitful concepts for the disciplines involved. The authors select problems showing representative traits concerning the frame up, the methods and the achievements rather than to present extended overviews.
The Mind and Brain are usually considered as one and the same nonlinear, complex dynamical system, in which information processing can be described with vector and tensor transformations and with attractors in multidimensional state spaces. Thus, an internal neurocognitive representation concept consists of a dynamical process which filters out statistical prototypes from the sensorial information in terms of coherent and adaptive n-dimensional vector fields. These prototypes serve as a basis for dynamic, probabilistic predictions or probabilistic hypotheses on prospective new data (see the recently introduced approach of "predictive coding" in neurophilosophy). Furthermore, the phenomenon of sensory and language cognition would thus be based on a multitude of self-regulatory complex dynamics of synchronous self-organization mechanisms, in other words, an emergent "flux equilibrium process" ("steady state") of the total collective and coherent neural activity resulting from the oscillatory actions of neuronal assemblies. In perception it is shown how sensory object informations, like the object color or the object form, can be dynamically related together or can be integrated to a neurally based representation of this perceptual object by means of a synchronization mechanism ("feature binding"). In language processing it is shown how semantic concepts and syntactic roles can be dynamically related together or can be integrated to neurally based systematic and compositional connectionist representations by means of a synchronization mechanism ("variable binding") solving the Fodor-Pylyshyn-Challenge. Since the systemtheoretical connectionism has succeeded in modeling the sensory objects in perception as well as systematic and compositional representations in language processing with this vector- and oscillation-based representation format, a new, convincing theory of neurocognition has been developed, which bridges the neuronal and the cognitive analysis level. The book describes how elementary neuronal information is combined in perception and language, so it becomes clear how the brain processes this information to enable basic cognitive performance of the humans.
