Issue 71, February 2016
bulletArtificial Intelligence
bulletInnovation: L2TOR: Using Robots to Help Immigrant Children Learn German
bulletBattling Robots to Compete in International RoboCup 2016 in Leipzig
bulletInterview with Reinhard Karger on the Future of Artificial Intelligence
bulletmedx/DX - A Digital Assistant to Support Physicians with Medical Diagnoses
bulletReimagining the Driving Experience: Mercedes-Benz F 015 Luxury in Motion - A Foreunner in the Mobility Revolution
Artificial Intelligence 
Can an algorithm commit a crime? Will a drone deliver your next holiday gift? Looking ahead, as we hand more decision-making over to intelligent technologies, will humanity face a threat? Or will the prospective benefits of artificial intelligence (AI) outweigh its potential costs?

Influential tech icons and renowned scientists, such as Elon Musk, Bill Gates, and Stephen Hawking, have all voiced their concerns regarding worst-case scenarios resulting from AI. To mitigate the risk of human consequences in the future, regulatory oversight on both a national and international level will be necessary.

With these fears taken into consideration, AI nonetheless offers exciting possibilities for us to re-imagine our living and working environments. One day, machines may learn, speak, reason, and even have a consciousness just like humans. They may even be able to teach one another - sharing observations and learning through AI to make them both individually and collectively smarter.

The following newsletter articles explore advances in artificial intelligence coming out of Germany - from robots used in the classroom to help refugee children learn German to intelligent software platforms designed to aid in faster, more accurate medical diagnoses. 




With hundreds of thousands of refugee children entering the German school system, teachers face unprecedented challenges. Teaching a child a second language individually is something that most classrooms cannot do. However, language acquisition is imperative to integration.

Could technical assistants be used to help prepare immigrant children for school and teach them a new language? L2TOR, pronounced 'el tutor' which stands for 'Second Language Tutoring Using Social Robots,' is a new project financed by the European Commission that is researching this very question.

One research group at Bielefeld University's Cluster of Excellence Cognitive Interaction Technology (CITEC) plans to provide tutoring systems with tablet PCs and the cute programmable robot 'Nao' to assist in language training. To do this, the researchers are developing modules that recognize a child's language abilities and motivation so that the robot can react individually to each child.

"We are investigating how interactive robots can be used to help teach children between the ages of four and six a second language," said Prof. Dr. Stefan Kopp, who heads the research group Social Cognitive Systems, which belongs to the Faculty of Technology and is part of CITEC. "We are working to provide children with the language abilities that they need for school."

Dr. Kirsten Bergmann, who works in Kopp's research group, explained, "It is important that the robot recognizes how the child being taught feels and whether he or she is frustrated or confused, for example." She continued, "We program the robot so that it can shape its interaction with the child so that he or she is being supported in the best way possible."

L2TOR just launched last month and will run for three years. Computer scientists, educators, and linguists are working together on this project as a consortium of five universities and two companies.

To read more in New Scientist's article "Robot language tutors to get kids up to speed before school," click here. For further questions, contact Prof. Dr. Stefan Kopp at [email protected] or Dr. Kirsten Bergmann at [email protected].

Source & Image: CITEC/Bielefeld University



Hundreds of robots will descend on Leipzig this summer to battle for the coveted title RoboCup 2016 World Champion. Soccer-playing humanoid robots will be at the center of the 20th RoboCup - the most prominent and largest competition for intelligent robots and one of the world's key technology events for artificial intelligence research and training. 

The competition, which aims to push advances in AI, will take place from June 30 to July 4, 2016, at the Leipzig exhibition grounds. Over 3,500 participants from 40 countries are expected to attend. The robots will compete against each other in eight different leagues, which vary in terms of how the robots are constructed and the type of program software they use.

RoboCup's unique global community - composed of several tens of thousands international participants - has set itself an ambitious goal: By 2050, it plans to field a team of fully autonomous humanoid robots that will defeat the reigning FIFA World Cup champions.

Aside from soccer, RoboCup competitions also demonstrate the use of intelligent robots for industry and business as well as for science and everyday life. This includes mobile robots for logistics and rescue robots for disaster response efforts.

In addition to the RoboCup Major League, an initiative called RoboCupJunior exists, which is dedicated to developing the next generation of scientists. The young innovators build and program their robots during school or in working groups to perform tasks like playing soccer or dancing. Workshops are also available for RoboCup visitors to gain insight into building and programming robots.

RoboCup is supported by a wide spectrum of political leaders and scientists. Its patrons include Prof. Dr. Johanna Wanka, German Federal Minister for Education and Research, and Martin Dulig, Saxony's Minister for the Economy, Labor, and Transportation. The City of Leipzig, the Free State of Saxony, and numerous companies and associations along with university and research institutions have also pledged their full support of the initiative.

To watch a video about the upcoming RoboCup, click here. To watch a clip of robots playing soccer in last year's semi-finals, click here.
 
Source & Image: Leipziger Messe GmbH

 



The German Research Center for Artificial Intelligence (Deutsches Forschungszentrum f�r K�nstliche Intelligenz, DFKI) is the world's largest non-profit contract research institute in the field of innovative software technology based on artificial intelligence (AI) methods. DFKI's research areas range from cyber-physical systems and multilingual technologies to embedded intelligence, augmented vision, and intelligent user interfaces. 

Reinhard Karger, the DFKI's Corporate Spokesperson and President of the German Association for Information and Knowledge (Deutsche Gesellschaft f�r Information und Wissen e.V., DGI), shared his thoughts with the GCRI on developments in AI. In his interview, Mr. Karger elaborates on what AI is and describes some of DFKI's current projects. He discusses what he foresees as the "next big thing" in AI and whether or not he believes AI poses a risk to jobs. He also addresses what, in his opinion, will be the most significant changes in the field over the next decade. To read the full interview, click here.

Mr. Karger received his master's in theoretical linguistics and later went on to work as a Research Assistant for a linguistics and logic project at IBM Deutschland as well as in the Department of Computational Linguistics and Phonetics at the Universit�t des Saarlandes in Saarbr�cken, Germany. Since joining the DFKI, Mr. Karger has held a variety of roles, including Project Manager for the long-term project Verbmobil funded by the German Federal Ministry of Research and Education (BMBF), Head of Corporate Communications, Head of the German Demonstration Center for Speech and Language Technology, and Corporate Spokesperson.

Image: Andr� Mail�nder   

 
No single doctor can consider all common and rare illnesses at the same time when making a diagnosis. Intelligent software systems, however, can help optimize care by drawing on millions of data points to identify patterns and think of things that doctors may forget. Smart software can help eliminate errors and avoid repetition of medical tests. 

medx/DX - a visual reasoning platform for physicians - is a software program designed to help doctors identify rare and particularly severe illnesses or cancer as early as possible. It supports clinical thinking for faster, safer diagnoses and better health outcomes. medx/DX combines a broad medical knowledge base, a fast, patient-adaptive reasoning algorithm, and an award-winning, patent-pending user interface.

However, unlike conventional artificial intelligence approaches, this system was not developed to replace human thinking. Rather, it aims to support a clinician's innate cognitive skills with the goal of ensuring and fostering cognitive excellence. Responsive data entry enables the rapid documentation of relevant information, a visual reasoning interface activates and supports clinical thinking and decision-making, and important diagnoses to consider are highlighted along with appropriate next steps in order to reduce risk.

medx/DX has been available in beta phase since July 2015. The company, which has offices in Munich, Berlin, and San Francisco, was founded in 2011 and has roughly 70 employees in Berlin. Its partners include the Parmenides Foundation, LMU, Durham University, University of Exeter, University of Freiburg, Klinik Dritter Orden Munich, TU Berlin, and Albion in London. medx is a 2014 Usability Award Winner and a 2015 Land of Ideas Ausgezeichnete Orte prize winner.

Source & Image: medx GmbH

Autonomous driving is one of the greatest innovations since the invention of the automobile. It relieves drivers of work and stress in situations in which driving is not enjoyable, such as stop-and-go traffic. Self-driving cars give people the freedom to do other activities while out on the road - adding quality time to their everyday lives.

With urbanization on the rise, in which space is at a premium and everything happening at a more hectic pace, people have a growing desire for privacy and a place to which they can retreat. The Mercedes-Benz F 015 Luxury in Motion - a futuristic, self-driving luxury sedan and visionary research vehicle - seeks to offer this refuge with its rolling lounge experience.

The sleek, pod-like vehicle offers a roomy cabin interior with space for up to four passengers. Its swiveling front seats can face forward like in a normal car or they can turn around to face the rear passengers to provide a more intimate setting. The seats also swivel 30 degrees outward when the doors open, making it easier for passengers to get in and out of the car's French doors.

Designed as a digital haven offering total connectivity, passengers can use the time spent in the vehicle for relaxing, communicating, or working. Six high-tech displays are installed around the interior for this purpose. Through gestures, eye-tracking, and touching high-resolution screens, passengers can intuitively interact with the networked vehicle for a complete immersive user experience.

On the exterior, large LED areas at the front and rear plus a forward-pointing laser projection system are responsible for visual communication with the vehicle's surroundings. On an acoustic level, sounds and specific voice instructions are part of the car's communication repertoire. The car's ability to interact with its surrounding environment makes the Mercedes-Benz F 015 Luxury in Motion a reliable partner on the road. 

In the future, with more free time gained from self-driving cars, the automobile will ultimately become a mobile living space rather than just a mere means of transport.

For exclusive insight from the engineering team of the Mercedes-Benz F 015 Luxury in Motion, click here for a video.  

Source & Image: Daimler Communications

MOSCOW        NEW DELHI       NEW YORK        S�O PAULO       TOKYO