Artificial intelligence needs guardrails


With the recent launch of the website as "Artificial Intelligence for the American People," AI will clearly be an integral part of our future. While some may still wonder, "what can AI do for us?," many more may be asking, "what can AI do to us?" given some recent tragic events. The crashes of the Boeing 737 MAXs and Uber and Tesla's self-driving car fatalities point to AI's unintended consequences and highlight how technologists as well as users of AI have both fallen short at making proper guardrails in deploying AI technology. People often think of AI as the panacea that will enable technology to solve our most pressing problems. In that way, AI brings to mind a seeming panacea of an earlier age: aspirin.

Facebook patents high-tech drone that uses kites to stay in the air for long periods of time

Daily Mail - Science & tech

Facebook has patented a high-tech drone that uses a unique apparatus to stay afloat. The filing, titled'Dual-kite aerial vehicle,' describes an unmanned aerial vehicle that is attached to two kites and can be flown at different altitudes. The kites allow the drone to remain in the air for an extended period of time'while consuming little or no fuel,' according to the patent. Facebook has patented a high-tech drone that uses a unique apparatus to stay afloat. The filing, 'Dual-kite aerial vehicle,' describes an unmanned aerial vehicle tethered to two kites The drone is attached to the two kites via a tether, which are each able to maintain flight at different altitudes.

Deep Reinforcement Learning for Event-Driven Multi-Agent Decision Processes Artificial Intelligence

The incorporation of macro-actions (temporally extended actions) into multi-agent decision problems has the potential to address the curse of dimensionality associated with such decision problems. Since macro-actions last for stochastic durations, multiple agents executing decentralized policies in cooperative environments must act asynchronously. We present an algorithm that modifies generalized advantage estimation for temporally extended actions, allowing a state-of-the-art policy optimization algorithm to optimize policies in Dec-POMDPs in which agents act asynchronously. We show that our algorithm is capable of learning optimal policies in two cooperative domains, one involving real-time bus holding control and one involving wildfire fighting with unmanned aircraft. Our algorithm works by framing problems as "event-driven decision processes," which are scenarios in which the sequence and timing of actions and events are random and governed by an underlying stochastic process. In addition to optimizing policies with continuous state and action spaces, our algorithm also facilitates the use of event-driven simulators, which do not require time to be discretized into time-steps. We demonstrate the benefit of using event-driven simulation in the context of multiple agents taking asynchronous actions. We show that fixed time-step simulation risks obfuscating the sequence in which closely separated events occur, adversely affecting the policies learned. In addition, we show that arbitrarily shrinking the time-step scales poorly with the number of agents.

Hierarchically Structured Meta-learning Machine Learning

In order to learn quickly with few samples, meta-learning utilizes prior knowledge learned from previous tasks. However, a critical challenge in meta-learning is task uncertainty and heterogeneity, which can not be handled via globally sharing knowledge among tasks. In this paper, based on gradient-based meta-learning, we propose a hierarchically structured meta-learning (HSML) algorithm that explicitly tailors the transferable knowledge to different clusters of tasks. Inspired by the way human beings organize knowledge, we resort to a hierarchical task clustering structure to cluster tasks. As a result, the proposed approach not only addresses the challenge via the knowledge customization to different clusters of tasks, but also preserves knowledge generalization among a cluster of similar tasks. To tackle the changing of task relationship, in addition, we extend the hierarchical structure to a continual learning environment. The experimental results show that our approach can achieve state-of-the-art performance in both toy-regression and few-shot image classification problems.

UAlbany Opens New Indoor Drone Lab - Unmanned Aerial


The University of Albany in Upstate New York recently unveiled a two-story, 1,700-square-foot drone lab. The College of Emergency Preparedness, Homeland Security and Cybersecurity (CEHC) hosted an open house last month to launch the lab, located in the basement of Page Hall at the university's downtown campus. The space, enclosed with netting and rubber flooring, offers a controlled indoor environment for unmanned aerial vehicle (UAV) flight training, along with land-based robotics research and educational opportunities. According to a press release from UAlbany, last summer, Brandon Behlendorf, an assistant professor at CEHC, was leaving his office in Richardson Hall when he stumbled upon an aging stairwell on the north corner of the second floor. Wondering where it led, he made his way down five stories to a locked door in the basement.

Future Tense Newsletter: An A.I. Cold War?


This is especially true of the tech sector, where some analysts liken the U.S. and China's heavy strategic investments in cybersecurity, quantum computing, 5G, and artificial intelligence to a digital arms race, one that, because of China's long-term positioning and access to vast amounts of data to train on, that country will win. But Anne-Marie Slaughter argues that when it comes to the world-shifting technology of artificial intelligence, the narrative isn't so simple. She explains why she is putting her money on the United States. Great power conflict isn't the only thing we at Future Tense have been fretting about this week. We've also been looking at digital privacy.

FAA Stands By Safety of Boeing 737 Max 8 After Deadly Crash

U.S. News

Earlier Tuesday, the Dallas Morning News reported that pilots made five complaints about the Boeing aircraft to federal authorities in the months leading up to Sunday's crash. The complaints, voluntarily made in the FAA's incident database, referenced problems with an autopilot system that occurred during the aircraft's ascent after takeoff, according to the Morning News. An FAA spokesman told the paper that such reports were filed directly to the National Aeronautics and Space Administration.

Drones retain their buzz at Japanese trade show, with industrial uses expected to bolster growth

The Japan Times

CHIBA - With the market for business-use unmanned aircraft looking promising in coming years, a large-scale drone expo that kicked off Wednesday showed more companies are eager to get involved with the trend. Companies ranging from the small to the powerful are showing off their business solutions using drones at Japan Drone, an annual exhibition at Makuhari Messe in Chiba that features more than 200 firms and runs until Friday. Telecom giant KDDI Corp. is showcasing its "smart" drone platform connected to KDDI's mobile communication networks across the country, which allows a drone to navigate a wider swath of territory via remote control. "One merit of using our service is that drones can be remote controlled through our communication networks anywhere in Japan, unlike most drones exhibited at this event, which tap Wi-Fi networks with limited coverage," said So Yamazaki, a KDDI official. KDDI will launch the service to corporate customers in June and lists surveillance, inspection, land survey and analysis as the envisioned applications.

Ordnance Survey launch a solar-powered drone that can fly for 90 days at a time

Daily Mail

Ordnance survey has unveiled a solar-powered drone that is capable of flying for 90 days at a time without needing to come back to Earth and will be used to provide higher quality images of Earth. It will circle at approximately 67,000 ft (20,400m) above the ground and snap images to sell to organisations and businesses. First tests of the Astigan unmanned aerial vehicle are scheduled to take place before the end of 2019. Ordnance Survey is the majority stakeholder in Astigan, a firm based in Bridgwater, Somerset. The company works in the same factory that was once home to Facebook's Aquila internet drone project.

Eurodrone Seeks Collision-Avoidance Feature – UAS VISION


The ministry's new missive to lawmakers came in response to a request made late last year by Andrej Hunko and fellow Linke members of parliament about the status of the Eurodrone. The questioners appear especially curious about IABG's role in helping to define key performance parameters for the aircraft, particularly related to airworthiness in civilian airspace. According to the ministry's letter, the contractor recommended outfitting the drone with a twin-turboprop propulsion system, a design feature that the developer nations ultimately adopted.