MRC aims to bring together researchers and practitioners from different communities, both industry and academia, to study, understand, and explore issues surrounding context and to share problems, techniques and solutions across a broad range of areas.
The workshop covers different understandings of what context is, a variety of approaches to automatically learn about context from data, and different approaches to modelling context. MRC is also concerned with varying mechanisms and techniques for reasoning with context, storage of contextual information, effective ways to retrieve it, and methods for enabling integration of context and application knowledge.
MRC invites papers on all aspects of context, from theoretical approaches over modelling, reasoning, and learning to reports on applications. This year our focus is on personalization, autonomy and privacy in relation to context. We explicitly invite contributions from other fields of study in order to further trans- and interdisciplinary approaches and further the integration of discipline specific knowledge into AI research.
We plan to publish the papers in accompanying online proceedings. Authors of (a subset of the) accepted papers may be invited to submit extended versions for inclusion in a special journal issue on contextualised systems, if justified by the quantity and quality of submissions.
For autonomous systems, recognising contextual information is vital if the system is to exhibit behaviour that is appropriate for the situation at hand. At the same time, such systems might change contextual parameters that are relevant for human and non-human agents present. Therefore, it is important to be able to predict changes in context that are due to the actions of intelligent systems to avoid clashing with user needs and expectations.
In multi-agent systems, contextual information might not be shared between the different actors explicitly or upfront. Therefore, it is vital for intelligent agents to identify the different context configurations and adapt their own behaviour accordingly.
From a machine learning perspective, contextual information might have to be learned from data before a contextualised system is being implemented. In many cases, contextual configurations might change over time, and cannot fully be modelled in the design phase of a system, necessitating machine learning methods to be employed during runtime.
From a general AI perspective, one of the challenges is to integrate context with other types of knowledge as a major additional source for reasoning, decision-making, and adaptation and to form a coherent and versatile architecture. There is a common understanding that achieving desired behaviour from intelligent systems will depend on the ability to represent and manipulate information about a rich range of contextual factors.
These factors may include not only physical characteristics of the task environment, but, possibly more importantly, many other aspects including cognitive factors such as the knowledge states (of both the application and user) or emotions, and social factors such as networks, relations, roles, and hierarchies. This representation and reasoning problem presents research challenges to which methodologies derived from areas such as artificial intelligence, knowledge management, human-computer interaction, semiotics and psychology can contribute solutions.
Areas of interest include, but are not limited to:
Please check our pages with author guidelines and submission procedure information for details.
Last modified: Sunday, 2020-01-12 19:28 UTC.
University of Hildesheim