KI - Künstliche Intelligenz

German Journal on Artificial Intelligence - Organ des Fachbereichs "Künstliche Intelligenz" der Gesellschaft für Informatik e.V.

ISSN: 0933-1875 (Print) 1610-1987 (Online)

Description

The Scientific journal "KI – Künstliche Intelligenz" is the official journal of the division for artificial intelligence within the "Gesellschaft für Informatik e.V." (GI) – the German Informatics Society – with contributions from throughout the field of artificial intelligence. The journal presents all relevant aspects of artificial intelligence – the fundamentals and tools, their use and adaptation for scientific purposes, and applications which are implemented using AI methods – and thus provides the reader with the latest developments in and well-founded background information on all relevant aspects of artificial intelligence. For all members of the AI community the journal provides quick access to current topics in the field and promotes vital interdisciplinary interchange.

Preview:

Answer Set Programming Unleashed

 

Answer Set Programming (ASP) has become a popular paradigm for Knowledge Representation and Reasoning (KRR), in particular, when it comes to solving knowledge-intense combinatorial (optimization) problems. The growing popularity of ASP in research and application domains rests upon the following pillars. First, ASP builds upon a simple yet rich modelling language with clear semantics that offers, for instance, cardinality and weight constraints as well as means to express multi-objective optimization functions. Second, all these constructs are well supported by highly performant solving technology leading to seamless support of such constraints along with sophisticated optimization algorithms. Finally, a primary asset of ASP is its versatility, arguably elicited by its roots in KRR and AI: ASP offers complex reasoning modes for enumerating, intersecting, or unioning solutions, as well as combinations thereof, e.g., intersecting all optimal solutions.

ASP can be looked at from different perspectives. For one, it can be seen as the computational embodiment of nonmonotonic reasoning. Similarly, it can be regarded as an extension of propositional logic and its solving machinery with closed world reasoning. For another, it can be viewed as an extension of database systems with possibly recursive rules. And although its original semantics was proposed to capture logic programs, its logical foundations have meanwhile been traced back to constructive logics.

This particular combination of different paradigms along with the aforementioned versatility made ASP a successful tool in AI research with a wide range of applications in academia as well as industry. Starting with an introduction to the essentials of ASP and its logical foundations, the special issue includes several articles on salient application areas of ASP. This is accompanied with interviews reflecting its upbringing from the early days of AI to modern off-theshelf ASP engines. And last but not least, the special issue features several reports from the field.

 

Torsten Schaub

University of Potsdam

torsten@cs.uni-potsdam.de

Stefan Woltran

TU Wien

woltran@dbai.tuwien.ac.at

 

Trust in AI

Special Issue Guest Editors:

* Raymond Sheh, Curtin University

* Claude Sammut, The University of New South Wales

* Mihai Lazarescu, Curtin University

This special issue on Trust in AI is dedicated to exploring the issue of how to improve understanding of the appropriate level of trust in AI-based systems among AI practitioners, users and broader society.

AI-based systems are becoming increasingly pervasive in modern society. Their influence extends from movie recommendations to mine optimisation, from border security to financial securities, from driver assistance to medical assessments. All of these applications bring with them different requirements in relation to, and definitions of, trust.

We welcome high quality submissions on issues of Trust in AI systems including, but not limited to:

* Trust in their level of performance - End to End Real World Performance Testing for AI Systems.

* Trust that they're making decisions for the correct reasons - Explainable AI (XAI).

* Trust in our ability to correct errors - Repairable AI and Lifelong Learning.

* Trust in their integrity - Cyber Security challenges faced by AI systems.

* Trust in our ability to oversee them - Oversight and Anomaly Detection for AI in Complex Tasks and Environments.

* Trust definitions, requirements and capabilities - Semantics and Definitions for Requirements Analysis around Trust in AI.

Applications of interest include, but are not limited to:

* Technologies for assisting human drivers, operators and other controllers.

* Collaborative and autonomous robots.

* Service, entertainment and other robots in broader society.

* Industrial optimisation.

* Medical sensing, smart sensors and diagnosis.

* Applied image and sensor understanding, recognition and analysis.

* Recommendation, diagnosis and decision support systems.

* Safety, security and mission critical systems across cyber, cyberphysical and physical domains.

* Big data analytics.

* AI in finance, trading and assignment of resources.

Formats that we consider include:

* Short and full length technical papers.

* Reports on research projects.

* Discussion pieces.

* Dissertation abstracts.

* Conference/workshop reports and survey papers.

* Industry papers (AI Market).

See Instructions for Authors (below) for the requirements for each of these formats.

 

Important Dates:

* 15 June 2018, deadline for submissions.

* 15 Jul 2018, notification of acceptance.

* 15 Aug 2018, final version due.

For comments, suggestions or requests please send email to Raymond Sheh: raymond.sheh@curtin.edu.au.

We look forward to your contribution to this special issue!

Browse Volumes & Issues