The lights are going out at malls across the United States with more than 20 major retail bankruptcies in 2017. As of today, store closures have skyrocketed to 7,000 doors throughout the nation, affecting such iconic brands as Toys R Us, Walgreens, Gap, Sam's Club, The Children's Place, Hallmark, Stride Ride, Aeropostale, Wet Seal, The Limited and Walmart. At the same time, investment in retail technology has never been higher, especially robots.A month after Walmart laid off close to 10,000 workers with the shuttering of Sam's Club, it announced a new partnership with Pittsburgh-based robot manufacturer, Bossa Nova. The mechatronics innovator will begin rolling out inventory auditing scanning bots to 50 Walmart locations. The machines will automate the tasks previously held by inventory associates by autonomously navigating around the store to check the shelf display, inventory position, and pricing of the big box's 200,000 items.

WASHINGTON – Dramatic new drone video of the Niger ambush that killed four American soldiers shows U.S. forces desperately trying to escape and fighting for their lives after friendly Nigerien forces mistook them for the enemy. It describes how the fleeing troops set up a quick defensive location on the edge of a swamp and -- thinking they were soon to die -- wrote messages home to their loved ones. The video, released by the Pentagon with explanatory narration, includes more than 10 minutes of drone footage, file tape and animation that wasn't made public last week when the military released a portion of the final report on the October attack. The video depicts for the first time the harrowing hours as troops held off their enemy and waited for rescue. There were 46 U.S. and Nigerien troops out on the initial mission in the West African nation, going after but failing to find a high-value militant, then collecting intelligence at a site where the insurgent had been.

Dramatic new drone video of the Niger ambush that killed four American soldiers shows U.S. forces desperately trying to escape, and fighting for their lives when friendly Nigerien forces mistook them for the enemy.

NEW YORK – AeroVironment Inc. was accused of trying to conceal that employees transported a drone rigged with explosives on a commercial flight and retaliating against a manager who told the government. In April 2015, AeroVironment workers traveled to Los Angeles from Salt Lake City on a Delta Air Lines Inc. There were about 230 civilian passengers aboard, the lawsuit states. The plaintiff, Mark Anderson, who oversaw security for the drone-maker's top-secret government programs, learned of the incident in May 2015, according to the complaint. After reporting it to the U.S. Department of Defense, he was reprimanded, stripped of his responsibilities and ultimately fired without severance, Anderson alleges.

This is part IV of our tour through the papers from the Re-coding Black Mirror workshop exploring future technology scenarios and their social and ethical implications. In 2016, the world witnessed the storming of social media by social bots spreading fake news during the US Presidential elections… researchers collected Twitter data over four weeks preceding the final ballot to estimate the magnitude of this phenomenon. Their results showed that social bots were behind 15% of all accounts and produced roughly 19% of all tweets… What would happen if social media were to get so contaminated by fake news that trustworthy information hardly reaches us anymore? Fake news and hoaxes have been around a long time, but the nature of social media pours fuel on the fire. Any user can create and relay content with little third-party filtering or fact-checking; many adults get their news on social media; and research has shown that people exposed to fake news tend to believe it.

The White House hosted an AI symposium on Thursday where reps from Google, Amazon, Facebook and the like met with academics to talk about how to best secure America's role as a world leader in artificial intelligence. They also announced the Select Committee on Artificial Intelligence, which will operate through the National Science and Technology Council and determine how best to leverage AI for American industry and market growth. And while it's great to see major investment in science and technology, it's also safe to assume that the Trump administration -- who once floated Rudy Guiliani as the country's chief of cybersecurity -- might not be, uh, up to the task of running this task force. The new AI task force aims to foster a free-market approach to technological advancement, according to AP. But history has shown time and time again through industries like airlines, big banks, and private healthcare companies that this, eh, doesn't always work out so well. Right now we're in a unique window of time where a lot of incredible technology and algorithms are emerging and influencing a massive portion of everyday life.

Sundar Pichai's demonstration of the company's new virtual-assistant technology, unveiled at the company's annual developer conference last week, was more unnerving than Pichai presumably intended it to be. Google Duplex, as the technology is called, represents a major leap forward in Silicon Valley's efforts to produce robots that sound like people. It can make phone calls to schedule appointments, say, or to reserve a table at a restaurant, using familiar human verbal tics and filler words--"uhm," "mmhmm," and "gotcha"--that make it eerily hard to tell that the voice on the other line is an artificial intelligence. To show the tech in action, Pichai played a recording of the Google Assistant device--Google's answer to Apple's Siri and Amazon's Alexa--calling and interacting with someone who was purportedly an employee at a hair salon to make an appointment. "What you're going to hear is the Google assistant actually calling a real salon to schedule an appointment for you," Pichai told the audience.

Introduction Lesson 1: Building Foundational Python Skills for Data Analytics Lesson 1 covers the Python tools commonly used for data analysis. Lesson 2: Analyzing Data Using Simulations and Resampling In Lesson 2 we apply the Python data analysis tools to building simulations and computing statistics. The techniques for resampling statistics are both powerful and easy to learn. They express big ideas with very little code. Lesson 3: Improving Reliability with MyPy and Typing Hinting In the first half of Lesson 3, we wade deeper into Python's tools for organizing and analyzing data.

Currently there is no standard way to identify how a dataset was created, and what characteristics, motivations, and potential skews it represents. To begin to address this issue, we propose the concept of a datasheet for datasets, a short document to accompany public datasets, commercial APIs, and pretrained models. The goal of this proposal is to enable better communication between dataset creators and users, and help the AI community move toward greater transparency and accountability. By analogy, in computer hardware, it has become industry standard to accompany everything from the simplest components (e.g., resistors), to the most complex microprocessor chips, with datasheets detailing standard operating characteristics, test results, recommended usage, and other information. We outline some of the questions a datasheet for datasets should answer. These questions focus on when, where, and how the training data was gathered, its recommended use cases, and, in the case of human-centric datasets, information regarding the subjects' demographics and consent as applicable. We develop prototypes of datasheets for two well-known datasets: Labeled Faces in The Wild~\cite{lfw} and the Pang \& Lee Polarity Dataset~\cite{polarity}.

Sequential learning of multiple tasks in artificial neural networks using gradient descent leads to catastrophic forgetting, whereby previously learned knowledge is erased during learning of new, disjoint knowledge. Here, we propose a new approach to sequential learning which leverages the recent discovery of adversarial examples. We use adversarial subspaces from previous tasks to enable learning of new tasks with less interference. We apply our method to sequentially learning to classify digits 0, 1, 2 (task 1), 4, 5, 6, (task 2), and 7, 8, 9 (task 3) in MNIST (disjoint MNIST task). We compare and combine our Adversarial Direction (AD) method with the recently proposed Elastic Weight Consolidation (EWC) method for sequential learning. We train each task for 20 epochs, which yields good initial performance (99.24% correct task 1 performance). After training task 2, and then task 3, both plain gradient descent (PGD) and EWC largely forget task 1 (task 1 accuracy 32.95% for PGD and 41.02% for EWC), while our combined approach (AD EWC) still achieves 94.53% correct on task 1. We obtain similar results with a much more difficult disjoint CIFAR10 task, which to our knowledge had not been attempted before (70.10% initial task 1 performance, 67.73% after learning tasks 2 and 3 for AD EWC, while PGD and EWC both fall to chance level). Our results suggest that AD EWC can provide better sequential learning performance than either PGD or EWC.