Leverhulme Centre for the Future of Intelligence. Computer vision for predicting physical urban change. Afghan girls allowed into US for robotics contest. http://ec2-35-176-202-215.eu-west-2.compute.amazonaws.com/news-briefing/.
DARPA is known for issuing big challenges. Still, the mission statement for its new Neural Engineering Systems Design program is a doozy: Make neural implants that can record high-fidelity signals from 1 million neurons.
Today’s best brain implants, like the experimental system that a paralyzed man used to control a robotic arm, record from just a few hundred neurons. Recording from 1 million neurons would provide a much richer signal that could be used to better control external devices such as wheelchairs, robots, and computer cursors.
The authors of this paper develop a computer vision method to measure changes in the physical appearances of neighborhoods from street-level imagery. They correlate the measured changes with neighborhood characteristics to determine which characteristics predict neighborhood improvement. They find that both education and population density predict improvements in neighborhood infrastructure, in support of theories of human capital agglomeration.
Neighborhoods with better initial appearances experience more substantial upgrading, as predicted by the tipping theory of urban change. Finally, they observe more improvement in neighborhoods closer to both city centers and other physically attractive neighborhoods, in agreement with the invasion theory of urban sociology.
Homeland Security Department spokesman David Lapan said the U.S. Citizenship and Immigration Services approved a State Department request for six girls from the war-torn country to be allowed in, along with their chaperone, so they can participate in the competition. The girls’ applications for U.S. visas had been denied twice.
The girls wanted to show the world that Afghans could also construct a hand-made robot and they had been deeply disappointed by the initial rejections. The team faced seemingly insurmountable obstacles to even get to a point where they could seek permission to attend. It took them six months to prepare, often working seven days a week, as they constructed a robot that sorts balls, has the ability to recognize orange and blue colors, and can move objects to put them in their correct places
Yesterday, Microsoft announced a new program called AI for Earth, aimed at putting the power of AI towards solving some of the biggest environmental challenges of our time. As a technology company with a deep commitment to sustainability, they understand that their responsibility extends beyond their own operations to innovating towards a healthier and better future more broadly.
Microsoft will offer access to cloud and AI computing resources, technology trainings and lighthouse projects – a $2 million commitment in this next fiscal year. And to lead this work, they’ve named Lucas Joppa, longtime Microsoft Research lead on computational ecology, to the role of chief environmental scientist.
Sam Charrington went to New York for the O’Reilly AI Conference. While there, he sat down with a bunch of great guests for interviews. The result is their O’Reilly AI NY podcast series—5 shows from the conference, which were all released this week, for your binge listening pleasure!
Google recently announced Gradient Ventures, a new venture fund with technical mentorship for early-stage startups focused on AI. Through Gradient, they’ll provide portfolio companies with capital, resources, and dedicated access to experts and bootcamps in AI.
Findo, an AI-focused smart search startup that’s striving to help people find information buried in cloud accounts, documents, and email attachments, has announced a new personal assistant designed to automatically create tasks and reminders based on the content of your emails.
Similar to Findo’s other AI-infused productivity tools, Yva.ai uses natural language processing (NLP) to dissect and understand the content of your emails, while leaning on machine learning to identify tasks issued to you within any of your emails.