1 – Research Reveals Robot Flaws are Key to Interacting with Humans
A new study reveals that humans are less likely to interact with robots if robots are programmed to be “too perfect.” The study was spearheaded by PhD researcher Mriganka Biswas and overseen by Dr. John Murray from the University of Lincoln’s School of Computer Science (UK), and presented at the Conference on Intelligent Robots and Systems (IROS) in Hamburg earlier this month. Mriganka said,
“…By developing these cognitive biases in the robots – and in turn making them as imperfect as humans – we have shown that flaws in their ‘characters’ help humans to understand, relate to and interact with the robots more easily.”
This is essential information, particularly as more robots continue to be developed for human interaction in the areas of support and caregiving.
(Read the full article on Phys.org)
2 – Automating Big-Data Analysis
MIT researchers’ “Data Science Machine” beat 615 out of 906 human teams in big-data analysis in three data science competitions. The teams’ goal is to take the human element out of data analysis, but only as a means of complementing and enhancing human capabilities. While humans still held their own about a third of the time, a key difference is the time element – where it took humans months to finish their analysis, the computer finished its interpretations between two to 12 hours. Max Kanter, whose master’s thesis in computer science is the basis for the machine’s design, will present research at next week’s IEEE International Conference on Data Science and Advanced Analytics.
(Read the full article on MIT News)
3 – Why Artificial Intelligence Should Read and Write Stories
Mark Ridel, associate professor at Georgia Tech School of Interactive Computing, has an end goal in mind for AI; he wants to create AI that have “narrative intelligence”, the ability to create and tell stories based on human reactions. This capability has many applications in the real world, from creating fictional fairy tales to creating simulate scenarios for real-world professionals (pilots, manufacturers, etc.) to participating in forensic investigations and create hypotheses of series of events based on evidence. Ridel and his fellow researchers are making progress in this area at the Georgia Institute of Technology Entertainment Intelligence Lab, where the Scheherazade system can already tell basic, fictional stories based on everyday information that is read to it.
(Read the full article on Huffington Post)
4 – Six Human Skills Facebook’s Artificial Intelligence Tech is Slowly Mastering
Facebook opened up its NYC-headquarters to media on October 14, giving a tour of its office space and a peek into many of its current and underway innovations, from Oculus to plane-sized drones. Of particular interest is its Artificial Intelligence team, which is focused primarily on language and machine vision. Most of the work is designated to research in six sub-branches: speaking; reading; translating; captioning; seeing; and refereeing (apparently, its ‘AI mind’ can tell the difference between various sports and knows the various rules).
(Read the full article on New York Business Journal)
5 – Publicis.Sapient Acquires Minority Stake in Artificial Intelligence Company Lucid
Publicis.Sapient – a division of parent company Publicis Groupe, which focuses on digital technologies – has acquired a minor stake in Lucid for an undisclosed amount. The investment will allow Publics.Sapient to enter in to the “big data” market with a new AI branch that uses tools to help aggregate data, looks for patterns and trends, and identify causal information behind those to improve clients’ marketing strategies. Global Head Josh Sutton remarked,
“We believe that artificial intelligence is going to be a transformative technology both on the marketing industry as well as the other industries that Publicis.Sapient is focused on.”
(Read the full article on The Wall Street Journal)