Most people spend the majority of their time at home, yet little is known about the air they breathe inside their houses. That’s why some atmospheric chemists are turning their attention toward indoor air, using tools developed for monitoring pollutants outside. By cataloguing compounds in indoor air, scientists could someday link them with health effects, according to an article in Chemical & Engineering News (C&EN), the weekly newsmagazine of the American Chemical Society.
Students and faculty at the Naval War College should begin “diving in” to artificial intelligence (AI), said the director of the Defense Department’s Joint Artificial Intelligence Center. Lt. Gen. Jack Shanahan, USAF, declared, “We need far more national security professionals who understand what this technology can do or, equally important, what it cannot do,” according to Navy officials.
Speaking to an assembly at the war college, Gen. Shanahan added, “On the other side of the equation, we desperately need more people who grasp the societal implications of the new technology, who are capable of looking at this new data-driven world through geopolitical, international relations, humanitarian and even philosophical lenses.”
The Australian Computer Society has launched a new Artificial Intelligence Hub from its Docklands, Melbourne premises.
The AI Hub will include 50 seats dedicated to artificial intelligence businesses operating onsite from Bay City Labs working with ACS to scale.
The space includes an eight panel multi-media wall to enable onsite tech scaleups to host stand-ups demonstrating their AI products and machine learning capabilities to customers, investors and other stakeholders.
Many members of the Chinese elite, even longtime advocates of market reform and economic opening, see a dark future for U.S.-China relations—and they are increasingly focused on America’s global financial hegemony as a long-term risk for their country. They’re indicating, subtly but unmistakably, that they see global finance as a rising theater of conflict, and are considering new ways for China to defend itself and even to retaliate.
Story Continued Below
An example is the economist Lou Jiwei, who has long been seen as a liberalizing force in China, an advocate of market reform and international openness. He served as finance minister, ran the country’s massive sovereign wealth fund, and has palled around with western economists since the 1980s. But recently he made a prediction that contained a startling phrase: At a forum in Beijing, according to reporting in the South China Morning Post, he said: “The next step in the frictions between China and the United States is a financial war ( jinrong zhan). The U.S. has been hijacked by nationalism and populism, so will do everything in its power to use bullying measures [and] long-arm jurisdiction.”
Facebook is spending six figures to fund a course on manipulated media and deepfakes for newsrooms, executives tell Axios. The course material has been developed by Reuters, and Facebook is funding its international expansion as a part of the Facebook Journalism Project.
IAC (InterActiveCorp.), an internet media holding company, agreed on Thursday to separate one of its largest subsidiaries, Match Group, which houses several dating apps including Tinder.
Why it matters: IAC’s stated goal has long been to build and/or acquire online services and to grow them to the point that they can be spun out independently, providing dual value to shareholders.
It was a sunny afternoon last month when my smartphone decided to ignore me. Well, it didn’t ignore my African American self, but it did ignore the carved face of a black man in a sculpture I was trying to photograph.
Instead, it bracketed the carving of a white man’s face to indicate that it was “seeing” him, while not bracketing the black face in the center of the frame. This problem of artificial intelligence having difficulty with black faces has been around for at least a decade. In addition to causing issues with commonplace activities, like taking pictures, it also raises a bigger question of how we guarantee equal opportunity across race in a world run on AI.
The Federal Trade Commission is challenging DNA sequencing giant Illumina’s proposed $1.2 billion purchase of rival Pacific Biosciences, alleging the acquisition would eliminate “a nascent competitive threat” to Illumina’s monopoly in deciphering genetic material.
“When a monopolist buys a potential rival, it can harm competition,” the deputy director of the FTC’s Bureau of Competition, Gail Levine, said Tuesday in a statement. “These deals help monopolists maintain power. That’s why we’re challenging this acquisition.”
Illumina has become dominant in the field of next-generation DNA sequencing by building machines that sequenced DNA more cheaply and rapidly than several competitors, many of which have gone out of business.
US lawmakers have reached an agreement that would fund gun-violence research for the first time in more than 20 years.
A spending bill introduced on 16 December includes US$25 million for studies on the issue, split evenly between the Centers for Disease Control and Prevention (CDC) and the National Institutes of Health (NIH). The House of Representatives and Senate are expected to approve the legislation this week, clearing the way for President Donald Trump to sign the bill into law.
“It’s a good start,” says Garen Wintemute, director of the Violence Prevention Research Program at the University of California, Davis, who has been studying gun violence for decades. “Violence-prevention policy should be guided by solid scientific evidence.”
A bipartisan budget agreement reached in the U.S. Congress will renew and strengthen the federal Open Textbook Pilot Grant Program with $7 million in funding for Fiscal Year 2020—a $2 million increase over previous years. Supported by advocates of higher education affordability, the renewal of this program for a third year is a resounding endorsement of the positive benefits that open textbooks can achieve for students. The funding is included in the Further Consolidated Appropriations Act, 2020, part of a $1.37 trillion budget package that Congress is expected to approve this week.
The Open Textbook Pilot offers grants to colleges and universities to implement programs that reduce the cost of textbooks through expanding the use of open educational resources—textbooks and other materials that are free for people everywhere to use and repurpose.
As a medical student at the University of Virginia, I remember encouraging my ER patients to accept the hospitalizations that they needed. When they, too, worried about the cost of care, I parroted a line I’d learned from residents and attendings throughout my training: “Don’t worry, the financial office will help you with that.”
But for these patients there was no help. After accepting care from UVA, the hospital system billed patients like mine and then sued them for an inability to pay, seizing the homes of some and bankrupting others.
Aiming to avoid a financial trauma of my own, I knew better than to consent to treatment without asking questions first.
When a patient representative showed up to our ER bay to let me know that my “insurance was verified,” I pressed her to clarify what that meant. Would I be billed for this admission? She didn’t know. No one did: not the billing office, nurses, case manager, pediatric ER fellow, or attending physician.
At the Tackling Climate Change workshop at this year’s NeurIPS conference, some of the top minds in machine learning came together to discuss the effects of climate change on life on Earth, how AI can tackle the urgent problem, and why and how the machine learning community should join the fight.
The panel included Yoshua Bengio, MILA director and University of Montreal professor; Jeff Dean, Google’s AI chief; Andrew Ng, cofounder of Google Brain and founder of Landing.ai; and Cornell University professor and Institute for Computational Sustainability director Carla Gomes.
Medical prescription records should be shielded from warrantless searches by federal agents cracking down on illegal use of opioids and other narcotics, New Hampshire officials are arguing in a First Circuit case.
The case is the first to test whether the Supreme Court’s decision last year in a cellphone privacy case, Carpenter v. United States, should apply to consumers’ prescription information that states collect and hold. Success for New Hampshire may encourage other states to thwart law enforcement access to the drug databases.
Madison, WI June 29-July 1, 2020, at University of Wisconsin. “Programme submissions are currently being requested for CarpentryCon, which will be held in Wisconsin from 29 June to 1 July 2020. Talented (and otherwise!) designers are also encouraged to submit their design ideas for the conference t-shirt.”
Seattle, WA July 6-8. “Submissions may range from early research prototypes to mature production-ready systems. Of particular interest are publicly available open-source or open-access systems. We additionally strongly encourage demonstrations of industrial systems that are technologically innovative given the current state of the art of theory and applied research in computational linguistics.” Deadline for submissions is January 31, 2020.
Recently, pre-trained models have achieved state-of-the-art results in various language understanding tasks, which indicates that pre-training on large-scale corpora may play a crucial role in natural language processing. Current pre-training procedures usually focus on training the model with several simple tasks to grasp the co-occurrence of words or sentences. However, besides co-occurring, there exists other valuable lexical, syntactic and semantic information in training corpora, such as named entity, semantic closeness and discourse relations. In order to extract to the fullest extent, the lexical, syntactic and semantic information from training corpora, we propose a continual pre-training framework named ERNIE 2.0 which builds and learns incrementally pre-training tasks through constant multi-task learning. Experimental results demonstrate that ERNIE 2.0 outperforms BERT and XLNet on 16 tasks including English tasks on GLUE benchmarks and several common tasks in Chinese. The source codes and pre-trained models have been released at this https URL.
Spotify Design blog, Mat Budelman and Mark Kizelshteyn
from
In a world where the product uniquely adapts to each user, we’ve found that creating deeply personalized products requires a new type of mindset and approach to design. Equipped with research about our users, and a deep understanding of business goals, we traditionally define detailed flows of a user’s journey, invent and refine interactions, and visually brand the experience—hopefully with some delight. But when working with Machine Learning at Spotify, we’re now tackling entirely new types of challenges.
Reflecting on a handful of projects at Spotify, we’ve come up with three principles we believe will help others design ML-powered experiences.
The Global AI Vibrancy Tool compares countries’ global activities, including both a cross-country perspective as well as an intra-country drill down. It is tempting to provide a single ranking of countries, but such rankings are notoriously tricky. Projecting complex, heterogeneous measures down to a single number (or even a small set of numbers) is fraught with methodological subtleties, and can be highly subjective or skewed. For now, we thought it more responsible and useful to provide a tool for the reader to set the parameters and obtain the perspective they find most relevant.