W17: 6th International Symposium on Attention in Cognitive Systems(ISACS'13)
Organizers: Lucas Paletta (JOANNEUM RESEARCH Forschungsgesellschaft mbH), Laurent Itti (University of Southern California), Björn Schuller (Technische Universität München)
The capacity to attend to the relevant has been part of Artificial Intelligence (AI) systems since the early days of the discipline. Currently, with respect to the design and computational modeling of artificial cognitive systems, selective attention has again become a focus of research, and one sees it important for the organization of behaviors, for control and interfacing between sensory and cognitive information processing, and for the understanding of individual and social cognition in humanoid artifacts. While visual cognition obviously plays a central role in human perception, findings from neuroscience and cognitive psychology have informed us on the perception-action nature of cognition. In particular, the embodiment in sensory-motor intelligence requires a continuous spatio-temporal interplay between interpretations from various perceptual modalities and the corresponding control of motor activities. In addition, the process of selecting information from the incoming sensory stream, in tune with contextual processing on a current task and global goals, becomes a challenging control issue within the viewpoint of focused attention. Seemingly attention systems must operate at many levels and not only at interfaces between a bottom-up driven world interpretation and top-down driven information selection. One may consider selective attention as part of the core of artificial cognitive systems. These insights have already produced paradigmatic changes in several AI-related disciplines, such as, in the design of behavior based robotics and the computational modeling of animats. Within the context of the engineering domain, the development of enabling technologies such as autonomous robotic systems, miniaturized mobile - even wearable - sensors, and ambient intelligence systems involves the real-time analysis of enormous quantities of data. These data have to be processed in an intelligent way to provide "on time delivery" of the required relevant information. Knowledge has to be applied about what needs to be attended to, and when, and what to do in a meaningful sequence, in correspondence with visual feedback.