Alle Publikationen
- <<
- <
- 1
2018
-
(2018) : Self-integrating Organic Control Systems: from Crayfish to Smart Homes In: ARCS: ARCS Workshop 2018; 31th International Conference on Architecture of Computing Systems: Berlin, Germany: Axel Springer SE, S. 1-8. Online verfügbar unter ARCS Workshop 2018; 31th International Conference on Architecture of Computing Systems
Abstract: Survival in complex environments, for both natural and artificial systems, requires behavioural adaptation to common changes and behavioural innovation to face the unexpected. The challenge here is to produce a vast variety of behaviours, each adapted to current circumstances, while relying on a limited amount of resources (e.g. sensors, controllers and actuators), within a 'suitable' time-frame. Drawing inspiration from neural and behavioural studies on crayfish, this position paper brings to the fore several key design features that enable organisms to address this challenge. It then proposes a similar design for artificial controllers, based on: i) an extensible set of reusable control units; and, ii) a goal-driven, context-sensitive (self-)integration process for assembling control units into a wide variety of integrated system controllers. Pre-integrated sub-controllers can also be merged, to improve efficiency while avoiding conflicts. The proposal is illustrated via a proof-of-concept implementation for the smart home, where users can add and remove goals and devices at runtime and the controller is adapted accordingly. This study brings us closer to our long-term objective of defining reusable methodologies and platforms for the development of self-* systems running in complex unpredictable environments, notably including smart homes, cities, vehicular networks and electrical grids, merged via the Internet of Things, and of People.
-
(2018): Virtual organizational design laboratory. Agent-based modeling of the co-evolution of social service delivery networks with population dynamics. In: Expert Systems with Applications 98, S. 189-204. DOI: 10.1016/j.eswa.2018.01.018
Abstract: This paper extends the concept of biological co-evolution to explain the performance and survival of one type of service organizations. It proposes that service delivery network design would benefit from complex adaptive systems (CAS) modeling approaches to recreate organizational phenomena driven by the interaction of the organization with its operating environment. This approach, paired with experimental design methods can serve as a virtual laboratory. We take the case of social service delivery (SSD) organizations that are structured as nonprofit organizations providing humanitarian assistance and relief services. These organizations often serve different geographical regions, thus, racial composition, migration patterns, and wealth of the populations served are factors that vary between locations. SSDs operate through service nodes (i.e., field offices, chapters, branches) in a network configuration. Therefore, the managerial decision of where to locate the field offices is an important one. An agent-based model to recreate agents' interactions as proxies of those exchanges occurring in real SSD settings is used. A series of validation experiments instill confidence that our model can be used as a virtual research laboratory. This paper contributes to the field of organizational design by testing a model able to recreate different policies that combined with different operating conditions impact the network over time and space. In addition, it provides experimental insights on what type of network configuration might provide a higher number of services delivered over time across the service network. The results can inform those defining the service system architecture looking to achieve SSD's goals considering the demographics of the markets served.
2017
-
(2017): An Evolutionary Transfer Reinforcement Learning Framework for Multiagent Systems. In: IEEE Transactions on Evolutionary Computation 21 (4), S. 601-615. DOI: 10.1109/TEVC.2017.2664665
Abstract: In this paper, we present an evolutionary transfer reinforcement learning framework (eTL) for developing intelligent agents capable of adapting to the dynamic environment of multiagent systems (MASs). Specifically, we take inspiration from Darwin's theory of natural selection and Universal Darwinism as the principal driving forces that govern the evolutionary knowledge transfer process. The essential backbone of our proposed eTL comprises several meme-inspired evolutionary mechanisms, namely meme representation, meme expression, meme assimilation, meme internal evolution, and meme external evolution. Our proposed approach constructs social selection mechanisms that are modeled after the principles of human learning to identify appropriate interacting partners. eTL also models the intrinsic parallelism of natural evolution and errors that are introduced due to the physiological limits of the agents' ability to perceive differences, so as to generate "growth" and "variation" of knowledge that agents have of the world, thus exhibiting higher adaptivity capabilities on solving complex problems. To verify the efficacy of the proposed paradigm, comprehensive investigations of the proposed eTL against existing state-of-the-art TL methods in MAS, are conducted on the "minefield navigation tasks" platform and the "Unreal Tournament 2004" first person shooter computer game, in which homogeneous and heterogeneous learning machines are considered.
2013
-
(2013) : Social Importance Dynamics. A Model for Culturally-Adaptive Agents In: Aylett, Ruth; Krenn, Brigitte; Pelachaud, Catherine; Shimodaira, Hiroshi (Hg.): Intelligent Virtual Agents: 13th International Conference IVA 2013 Proceedings: Berlin; Heidelberg: Springer, S. 325-338
DOI: https://doi.org/10.1007/978-3-642-40415-3_29 Abstract: The unwritten rules of human cultures greatly affect social behaviour and as such should be considered in the development of socially intelligent agents. So far, there has been a large focus on modeling cultural aspects related to non-verbal behaviour such as gaze or body posture. However, culture also dictates how we perceive and treat others from a relational perspective. Namely, what do we expect from others in different social situations and how much are we willing to do for others as well. In this article we present a culturally configurable model of such social dynamics. The aim is to facilitate the creation of agents with distinct cultural behaviour, which emerges from different parametrisations of the proposed model. The practical application of the model was tested in the development of an agent-based application for intercultural training, in which the model is responsible for driving the socio-cultural behaviour of the virtual agents.
Keywords: Adaptive Systeme, Disziplin, doppelter Treffer, Favoriten, Kultur, kulturadaptives System, Kulturspezifizität, Kulturtheorie, Mensch-Technik-Relationen (MTR), Modell, Modelle/Theorien, Modellierung, Realtechnik, social importance dynamics, Soziale Angemessenheit, Soziosensitive Systeme, soziosensitives System, Technik, Volltreffer 2011
-
(2011) : Etiquette Considerations For Adaptive Systems That Interrupt: Cost And Benefits In: Hayes, Caroline C.; Miller, Christopher Allan (Hg.): Human-computer etiquette: Cultural expectations and the design implications they place on computers and technology: Boca Raton, Fla.: CRC Press, S. 289-322
Abstract: It is not generally considered impolite to interrupt. However, etiquette is not simply about politeness but also appropriateness. Thus, while etiquette dictates that it is inappropriate, counterproductive, or even dangerous to interrupt someone with irrelevant information in the middle of a critical task requiring great concentration, it also allows that it is appropriate and necessary to interrupt with crucial information. For example, it is inappropriate to interrupt an airline pilot during landing with the baseball score, but entirely appropriate to interrupt with the information that the landing gear has malfunctioned.
Keywords: Adaptive Systeme, Dialog/Konversation/Rede, doppelter Treffer, Etikette, Favoriten, Human-Computer Etiquette, Interuption, Intervention/Unterbrechung etc., Mensch-Technik-Relationen (MTR), Mensch-Technische Systeme, Realtechnik, Relevanz des Grundes für eine Unterbrechung, Sozial Intelligente Agenten, Soziale Angemessenheit, Soziale Robotik, Soziosensitive Systeme, Technik, Volltreffer 2005
-
(2005): Nonlinearity, chaos, and complexity. The dynamics of natural and social systems. Oxford; New York: Oxford University Press. Online verfügbar unter http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&db=nlabk&AN=177022
Abstract: Covering a broad range of topics, this text provides a comprehensive survey of the modelling of chaotic dynamics and complexity in the natural and social sciences. Its attention to models in both the physical and social sciences and the detailed philosophical approach make this an unique text in the midst of many current books on chaos and complexity. Part 1 deals with the mathematical model as an instrument of investigation. The general meaning of modelling and, more specifically, questions concerning linear modelling are discussed. Part 2 deals with the theme of chaos and the origin of chaotic dynamics. Part 3 deals with the theme of complexity: a property of the systems and of their models which is intermediate between stability and chaos. Including an extensive index and bibliography along with numerous examples and simplified models, this is an ideal course text.
Keywords: Adaptive Systeme, Chaos, Chaotic behavior in systems, Dynamics, Dynamique, Im (wissenschaftlichen) Diskurs, Komplexitätsforschung, Künstliche Intelligenz, Mathematical models, Mathematik, Mensch-Technik-Relationen (MTR), Modèles mathématiques, Modell, Modellierung, Multi-Agenten-Systeme, Nonlinear theories, Ontologische Differenzierungen: Tier, Mensch, Maschine, Realtechnik, Reflexionen über MTR, social systems, Soziale Ordnung, Soziale Systeme, Systemtheorie, Technik, Théories non linéaires, Überblick 2001
-
(2001): Let's talk!. Socially intelligent agents for language conversation training. In: IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans 31 (5), S. 465-471. DOI: 10.1109/3468.952722
Abstract: This paper promotes socially intelligent animated agents for the pedagogical task of English conversation training for native speakers of Japanese. Since student-agent conversations are realized as role-playing interactions, strong requirements are imposed on the agents' affective and social abilities. As a novel feature, social role awareness is introduced to animated conversational agents, that are by now strong affective reasoners, but otherwise often lack the social competence observed in humans. In particular, humans may easily adjust their behavior depending on their respective role in a social setting, whereas their synthetic pendants tend to be driven mostly by emotions and personality. Our main contribution is the incorporation of a "social filter program" to mental models of animated agents. This program may qualify an agent's expression of its emotional state by the social context, thereby enhancing the agent's believability as a conversational partner. Our implemented system is web-based and demonstrates socially aware animated agents in a virtual coffee shop environment. An experiment with our conversation system shows that users consider socially aware agents as more natural than agents that violate conventional practices.
Keywords: Adaptive Systeme, Agent, Computerwissenschaft, doppelter Treffer, Emotions, emotionssensitiv, Favoriten, Ingenieurswissenschaft, Interaktion, Kommunikation, Künstliche Intelligenz, Linguistik, Mensch-Technik-Relationen (MTR), Modelle/Theorien, Realtechnik, social filter program, social filter program for animated agents, Soziale Angemessenheit, soziale intelligente Agenten, sozialer Kontext, Sozialkompetenz, Soziosensitive Systeme, Spracherwerb, Teaching, Sprachverstehen, Technik, training in social competences, Volltreffer
- <<
- <
- 1