Refine
Year of publication
Document Type
Way of publication
- Open Access (2)
Keywords
- machine learning (2)
- Biosemiotik (1)
- Cognitive system (1)
- Cognitive systems (1)
- Computational cognition (1)
- Computational linguistics (1)
- Fock space (1)
- Künstliche Intelligenz (1)
- Maschinensemiotik (1)
- Matlab (1)
Institute
Semantic Dialogue Modeling
(2012)
Little Drop of Mulligatawny Soup, Miss Sophie? Automatic Speech Understanding provided by Petri Nets
(2017)
We report on research conducted as part of the Universal Cognitive User Interface (UCUI) project, which aims at developing a universal, autarkic module for intuitive interaction with technical devices. First, we present an empirical study of image schemas as basic building blocks of human knowledge. Image schemas have been studied extensively in cognitive linguistics, but insufficiently in the context of human-computer-interaction (HCI). Some image
schemas are developed early at pre-verbal stages (e.g., up-down) and may, thus, exert greater influence on human knowledge than later developed image schemas (e.g., centre-periphery). To investigate this for HCI contexts, we applied a speech interaction task using a Wizard of Oz paradigm. Our results show that users apply early image schemas more frequently than late image schemas. They should, therefore, be given preference in interface designs. In the second
part of this contribution we therefore focus on the appropriate representation and processing of semantics. We introduce novel theoretical work including feature-values-relations and Petri net
transducers, and discuss their impact on behaviour control of cognitive systems. In addition, we illustrate some details of the implementation regarding learning strategies and the graphical
user interface.
Vector symbolic architectures (VSA) are a viable approach for the hyperdimensional representation of symbolic data, such as documents, syntactic structures, or semantic frames. We present a rigorous mathematical framework for the representation of phrase structure trees and parse trees of context-free grammars (CFG) in Fock space, i.e. infinite-dimensional Hilbert space as being used in quantum field theory. We define a novel normal form for CFG by means of term algebras. Using a recently developed software toolbox, called FockBox, we construct Fock space representations for the trees built up by a CFG left-corner (LC) parser. We prove a universal representation theorem for CFG term algebras in Fock space and illustrate our findings through a low-dimensional principal component projection of the LC parser state. Our approach could leverage the development of VSA for explainable artificial intelligence (XAI) by means of hyperdimensional deep neural computation.
Cognitive agents that act independently and solve problems in their environment on behalf of a user are referred to as autonomous. In order to increase the degree of autonomy, advanced cognitive architectures also contain higher-level psychological modules with which needs and motives of the agent are also taken into account and with which the behavior of the agent can be controlled. Regardless of the level of autonomy, successful behavior is based on interacting with the environment and being able to communicate with other agents or users. The agent can use these skills to learn a truthful knowledge model of the environment and thus predict the consequences of its own actions. For this purpose, the symbolic information received during the interaction and communication must be converted into representational data structures so that they can be stored in the knowledge model, processed logically and retrieved from there. Here, we firstly outline a grammar-based transformation mechanism that unifies the description of physical interaction and linguistic communication and on which the language acquisition is based. Specifically, we use minimalist grammar (MG) for this aim, which is a recent computational implementation of generative linguistics. In order to develop proper cognitive information and communication technologies, we are using utterance meaning transducers (UMT) that are based on semantic parsers and a mental lexicon, comprising syntactic and semantic features of the language under consideration. This lexicon must be acquired by a cognitive agent during interaction with its users. To this aim we outline a reinforcement learning algorithm for the acquisition of syntax and semantics of English utterances. English declarative sentences are presented to the agent by a teacher in form of utterance meaning pairs (UMP) where the meanings are encoded as formulas of predicate logic. Since MG codifies universal linguistic competence through inference rules, thereby separating innate linguistic knowledge from the contingently acquired lexicon, our approach unifies generative grammar and reinforcement learning, hence potentially resolving the still pending Chomsky-Skinner controversy.
btuktuminglear, btuktsptech, btuktcogsys
Machine Semiotics
(2023)
Recognizing a basic difference between the semiotics of humans and machines presents a possibility to overcome the shortcomings of current speech assistive devices. For the machine, the meaning of a (human) utterance is defined by its own scope of actions. Machines, thus, do not need to understand the conventional meaning of an utterance. Rather, they draw conversational implicatures in the sense of (neo-)Gricean pragmatics. For speech assistive devices, the learning of machine-specific meanings of human utterances, i.e. the fossilization of conversational implicatures into conventionalized ones by trial and error through lexicalization appears to be sufficient. Using the quite trivial example of a cognitive heating | device, we show that — based on dynamic semantics — this process can be formalized as the reinforcement learning of utterance-meaning pairs (UMP).
We present a model, inspired by quantum field theory, of the so-called inner stage of technical cognitive agents. The inner stage represents all knowledge of the agent. It allows for planning of actions and for higher cognitive functions like coping and fantasy. By the example of a cognitive mouse agent living in a maze wold, we discuss learning, action planning, and attention in a fully deterministic setting and assuming a totally observable world. We explain the relevance of our approach to cognitive infocommunications.
In this contribution, we want to summarize recent development steps of theembeddedcognitive user interfaceUCUI, which enables auser-adaptive scenario inhuman-machine or even human-robot interactions by considering sophisticated cognitive andsemantic modelling.The interface prototype is developed by differentGerman institutes and companies with their steering teams at Fraunhofer IKTS and Brandenburg University of Technology. The interface prototype is able to communicate with users viaspeech andgesture recognition, speech synthesis and a touch display. The device includes an autarkic semantic processing and beyond a cognitive behavior control, which supportsan intuitive interaction to controldifferent kindsof electronic devices,e.g. in a smart home environment or in interactive respectively collaborative robotics.Contraryto available speech assistance systems such as Amazon Echo or Google Home, the introduced cognitive user interface UCUIensuresthe user privacy by processing all necessary informationwithout any network access ofthe interface device.
We present a Matlab toolbox, called “FockBox”, handling
Fock spaces and objects associated with Fock spaces: scalars, ket and
bra vectors, and linear operators. We give brief application examples
from computational linguistics, semantics processing, and quantum logic,demonstrating the use of the toolbox.
Das Fachgebiet der Kognitiven Technischen Systeme zeichnet sich durch einen hohen Grad an Interdisziplinarität (z. B. Kenntnisse auf den Gebieten Biologie, Psychologie, Informatik und den Ingenieurwissenschaften) aus. Nach wie vor besteht Bedarf an einer methodischen Darstellung des Fachgebietes, bei der die theoretische Durchdringung von Zusammenhängen zwischen den verschiedenen Wissensgebieten zusätzlich durch Anschaulichkeit unterstützt wird. Die vorliegende Arbeit stellt einen Fortschrittsbericht zur Realisierung eines Forschungs- und Experimentiersystems dar, mit dem wir dieses Anliegen unterstützen und über das wir erstmals konzeptionell in [1] berichtet haben. In diesem Beitrag folgen wir einem integrativen Ansatz zur Entwicklung einer kognitiven Architektur, mit der unter Verwendung repräsentationaler Datenstrukturen adaptives Verhalten auf verschiedenen Zeitskalen sowie zwei wichtige Verhaltensprogramme für das Problemlösen (Objektfindung, Exploration) auf der gemeinsamen Grundlage von Markov-Entscheidungsprozessen umgesetzt werden. Mit einem Kurzbericht zum Entwicklungsstand der physikalischen Experimentierumgebung und einer Zusammenfassung der bislang erreichten Ergebnisse beschließen wir den diesjährigen Beitrag.
In diesem Text wird das in „Maschinensemiotik“ [12] vorgestellte Verfahren zur Verknüpfung von Äußerungen mit deren Bedeutung weiterentwickelt und in Form von Algebraischen Petri-Netzen [20] modelliert. Gleichzeitig wird der Formalismus der Petri-Netze nebenbei eingeführt, wobei die Vorstellung nicht auf formale Vollständigkeit abzielt, sondern nur das für die Anwendung nötige Wissen Erwähnung findet. Die Modellierung umfasst dabei sowohl die Systemkomponente, die Verknüpfungen zwischen sprachlichen Äußerungen und Handlungen herstellt – das Lernermodell –, als auch die Komponente, die sprachliche Äußerungen tätigt, bis die gewünschte Verknüpfung hergestellt wurde – das Lehrermodell samt Lernzielkontrolle.
Die Umsetzung des Modells per Software [19] erlaubt die weitere Diskussion anhand einer Simulation mit einigen Testdaten. Eine ausführliche mathematische Analyse zeigt beweisbare Eigenschaften des Modells auf. Knappe Vergleiche mit der kybernetischen Didaktik nach von Cube [26] sowie der Suggestopädie Lozanovs [16] ordnen das Modell darüber hinaus noch in einen pädagogischen Kontext ein.
Zunächst möchten wir in das Thema einführen, indem wir kurz auf
gegenwärtige Probleme und zukünftige Möglichkeiten von Sprachassistenten eingehen. Es folgen grundlegende Überlegungen zur Semantik und zwei fundamentale Erkenntnisse aus der Biosemiotik und dem Konstruktivismus. Danach fokussieren wir uns auf den Spracherwerbsprozess. Hierbei kommt ein Petrinetz zum Einsatz, das einem elementaren, bereits in früheren Arbeiten vorgestellten, Verstärkungsalgorithmus folgt. Dem schließt sich eine Demonstration des Spracherwerbs anhand eines Beispiels aus dem Smart Car-Bereich an.
btuktcogsys
Bereits in der Frühphase der Kybernetik stellte C. E. Shannon ein Labyrinthexperiment zur Problemlösung vor, das sich an der mythologischen Figur des Theseus orientierte. Seine Lösung kommt dabei ohne ein Bedeutungskonzept aus. Eine Problemanalyse aus psychologischer Sicht macht allerdings deutlich, dass bei diesem Experiment wesentliche kognitive Aspekte nicht berücksichtigt wurden. Um diese Aspekte herauszustellen, befassen wir uns kurz mit den mythologischen und historischen Hintergründen des Labyrinthproblems und erläutern anschließend, warum wir über Shannons Modellvorstellungen hinausgehen müssen und Verfahren benötigen, mit denen wir zu semantischen Datenträgern gelangen können. Zur Veranschaulichung unserer Methode modifizieren wir die Problemstellung, in dem wir "Theseus" die Befriedigung der Bedürfnisse Hunger und Durst ermöglichen. Die Grundlage des Verfahrens ist die Festlegung von Zielmerkmalen und daran anschließend die Identifikation von lösungsrelevanten Steuermerkmalen, welche "Theseus" zur Problemlösung nutzen kann. Wir verwenden dabei den mathematischen Formalismus der Quantenmechanik, da er uns die Ableitung von semantischen Strukturen auf eine sehr elegante Weise erlaubt. Zunächst exploriert "Theuseus" die Umgebung und erhält eine kompakte Repräsentation in Form eines Weltvektors in einem Hilbertraum, der durch Tensorverknüpfungen verschiedener Merkmalsräume entsteht. Die Kombination dieser Merkmalsräume führt auf die mathematische Struktur eines Verbandes. Ausgehend vom gefundenen Weltvektor stellen wir ein Verfahren vor, mit dem "Theseus" Unterstrukturen des Verbandes bestimmen kann, welche semantischen Strukturen entsprechen und anschließend für die Problemlösung verwendet werden können. Wir demonstrieren das Verfahren an verschiedenen Weltkonfigurationen. Den dabei abgeleiteten semantischen Strukturen kann man entnehmen, welche Merkmale berücksichtigt werden müssen, um die Zielmerkmale einer gegebenen Konfiguration manipulieren zu können.
In this paper we give instructions on how to write a minimalist grammar (MG). In order to present the instructions as an algorithm, we use a variant of context free grammars (CFG) as an input format. We can exclude overgeneration, if the CFG has no recursion, i.e. no non-terminal can (indirectly) derive to a right-hand side containing itself. The constructed MGs utilize licensors/-ees as a special way of exception handling. A CFG format for a derivation A_eats_B↦∗peter_eats_apples, where A and B generate noun phrases, normally leads to overgeneration, e.\,g., i_eats_apples. In order to avoid overgeneration, a CFG would need many non-terminal symbols and rules, that mainly produce the same word, just to handle exceptions. In our MGs however, we can summarize CFG rules that produce the same word in one item and handle exceptions by a proper distribution of licensees/-ors. The difficulty with this technique is that in most generations the majority of licensees/-ors is not needed, but still has to be triggered somehow. We solve this problem with ϵ-items called \emph{adapters}.
The concept of intelligent agents is—roughly speaking—based on an architecture and a set of behavioral programs that primarily serve to solve problems autonomously.
Increasing the degree of autonomy and improving cognitive performance, which can be assessed using cognitive and behavioral tests, are two important research trends. The degree of autonomy can be increased using higher-level psychological modules with which needs and motives are taken into account. In our approach we integrate these modules in architecture for an embodied, enactive multi-agent system, such that distributed problem solutions can be achieved. Furthermore, after uncovering some weaknesses in the cognitive performance of traditionally designed agents, we focus on two major aspects. On the one hand, the knowledge processing of cognitive agents is based on logical formalisms, which have deficiencies in the representation and processing of incomplete or uncertain knowledge. On the other hand, in order to fully understand the performance of cognitive agents, explanations at the symbolic and subsymbolic levels are required. Both aspects can be addressed by quantum-inspired cognitive agents. To investigate this approach, we consider two tasks in the sphere of Shannon's famous mouse-maze problem: namely classifying target objects and ontology inference. First, the classification of an unknown target object in the mouse-maze, such as cheese, water, and bacon, is based on sensory data that measure characteristics such as odor, color, shape, or nature. For an intelligent agent, we need a classifier with good prediction accuracy and explanatory power on a symbolic level. Boolean logic classifiers do work on a symbolic level but are not adequate for dealing with continuous data. Therefore, we demonstrate and evaluate a quantum-logic-inspired classifier in comparison to Boolean-logic-based classifiers. Second, ontology inference is iteratively achieved by a quantum-inspired agent through maze exploration. This requires the agent to be able to manipulate its own state by performing actions and by collecting sensory data during perception. We suggest an algebraic approach where both kinds of behaviors are uniquely described by quantum operators. The agent's state space is then iteratively constructed by carrying out unitary action operators, while Hermitian perception operators act as observables on quantum eigenstates. As a result, an ontology emerges as the simultaneous solution of the respective eigenvalue equations.
Tags: btuktqiai; btuktcogsys; btukttheseus; btuktqtheseus
Speech-controlled user interfaces facilitate the operation of devices and household functions to laymen. State-of-the-art language technology scans the acoustically analyzed speech signal for relevant keywords that are subsequently inserted into semantic slots to interpret the user's intent. In order to develop proper cognitive information and communication technologies, simple slot-filling should be replaced by utterance meaning transducers (UMT) that are based on semantic parsers and a mental lexicon, comprising syntactic, phonetic and semantic features of the language under consideration. This lexicon must be acquired by a cognitive agent during interaction with its users. We outline a reinforcement learning algorithm for the acquisition of the syntactic morphology and arithmetic semantics of English numerals, based on minimalist grammar (MG), a recent computational implementation of generative linguistics. Number words are presented to the agent by a teacher in form of utterance meaning pairs (UMP) where the meanings are encoded as arithmetic terms. Since MG encodes universal linguistic competence through inference rules, thereby separating innate linguistic knowledge from the contingently acquired lexicon, our approach unifies generative grammar and reinforcement learning, hence potentially resolving the still pending Chomsky-Skinner controversy.
ucuikt2015