Dear readers,
While all these topics are very fascinating from a technical perspective—and this is the main focus of KI—, the consequences of the massive digitalization of our everyday life may not be ignored when we talk about the pros and cons of all these current technical developments.
At the end of 2015, in their “Digital Manifest: Digitale Demokratie statt Datendiktatur” (http://www.spektrum.de/pdf/digital-manifest/1376682) Dirk Helbing, Bruno S. Frey, Gerd Gigerenzer, Ernst Hafen, Michael Hagner, Yvonne Hofstetter, Jeroen van den Hoven, Roberto Zicari, and Andrej Zwitter pointed out the omnipresent collection of data that is partially enabled by smart environments may lead to severe problems for individuals and societies as a whole: Personalisation, first understood part of a solution to information overload, leads to extreme data filtering known as the filter bubble: humans who rely on digital assistance systems loose control over the information they obtain and—more importantly—the information they do not obtain as well as the ways in which information is presented to them. Decisions made by systems trained by machine learning algorithms are—from the user’s point of view—biased by the training sample which itself may be biased by the intentions and goals of those of built the sample. Therefore, the main ethical and political issue is that of how our democratic value system can be preserved given recent and future developments in fields as such big data, computational intelligence, and machine learning. Mostly importantly, we all have to remain intellectual individuals able to judge and criticise what happens around us. Another aspect is that all the data collected and being collected in the future should be stored transparently and under public control. This may lead to political and economic conflicts with private institutions interested in having many data under control. However, this debate has to be started and continued in parallel with technical developments, as there will never be technical solutions that are capable of solving these conflicts. We all will have to discuss them, find new legal and political rules what is allowed with data and what is forbidden. If we can conduct this debate constructively and cooperatively, then technologically smart environments will also be smart in everyday life for all of us.
So participate in the debate and, in order to understand problems and challenges, read this issue of KI and enjoy it!
Bernd Ludwig
Forthcoming Special Issues
1 Semantic Interpretation of Multi-Modal Human-Behaviour Data
-
Visuo-spatial imagery.
-
Movement and interaction data.
-
Neurophysiological and other human behaviour data.
Proposed foundational methods will, for instance, present the development of human-centred technologies and cognitive interaction systems aimed at assistance and empowerment, e.g. in everyday life and professional problem solving and creativity. This call particularly emphasises systematically formalised integrative AI methods and tools (e.g., combining reasoning and learning) that enable declarative modelling, reasoning and query answering, relational learning, embodied grounding and simulation etc. Broadly, the role of declarative abstraction, knowledge representation and reasoning, and neural-symbolic learning and inference from multi-modal sensory data is highly welcome. For details, please refer to the full Call for Papers at http://hcc.uni-bremen.de/calls/SpecialIssue-KI.pdf, or contact one of the guest editors:
Prof. Dr. Mehul Bhatt
University of Bremen, Germany
bhatt@uni-bremen.de
Prof. Dr. Kristian Kersting
Technical University of Dortmund, Germany
kristian.kersting@cs.tu-dortmund.de
2 Algorithmic Challenges and Opportunities of Big Data
Computer systems pervade all parts of human activity and acquire, process, and exchange data at a rapidly increasing pace. As a consequence, we live in a big data world where information is accumulating at an exponential rate and complexity, and often the real problem has shifted from collecting enough data to dealing with its impetuous growth and abundance when going through it to mine relevant or pertinent information. In fact, we often face poor scale-up behaviour from algorithms that have been designed based on models of computation that are no longer realistic for big data. This implies challenges like algorithmic exploitation of parallelism (multicores, GPUs, parallel and distributed systems, etc.), handling external and outsourced memory as well as memory-hierarchies (clouds, distributed storage systems, hard-disks, flash-memory, etc.), dealing with large scale dynamic data updates and streams, compressing and processing compressed data, approximation and online processing respectively mining under resource constraints, increasing the robustness of computations (e.g., concerning data faults, inaccuracies, or attacks) or reducing the consumption of energy by algorithmic measures and learning. Only then big data will truly open up unprecedented opportunities for both scientific discoveries and commercial exploitation in Artificial Intelligence, Geoscience, Social Web, Finance, e-Commerce, Health Care, Environment and Climate, Physics and Astronomy, Chemistry, Agriculture, Life Sciences and Digital Libraries, among other domains.
The aim of the special issue is to collect overview articles on important state-of-the-art algorithmic foundations and applications, as well as articles on emerging trends for the future of big data.
If you are interested in contributing to this special issue, please contact one of the guest editors before the submission deadline of March 1st, 2017:
Prof. Dr. Kristian Kersting
Technical University of Dortmund, Germany
Fakultät für Informatik
kristian.kerting@cs.tu-dortmund.de
Prof. Dr. Ulrich Meyer
Goethe Universität Frankfurt am Main, Germany
Institut für Informatik
umeyer@cs.uni-frankfurt.de