Artificial intelligence is here, and it's impacting our lives in real ways -- whether it's the Alexa smart speaker on our nightstand, online customer service chatbots, or the smart replies Google drafts for our emails. But so far, the tech's development has outpaced regulation. Now, government agencies are increasingly encountering AI-based tools, and they must figure out how to evaluate them. Take the Food and Drug Administration, which greenlights new medical products: It needs to review and approve new health care products that boast AI-capabilities -- like this one that promises to detect eye problems related to diabetes -- before they're sold to us. Or consider the Equal Employment Opportunity Commission, which investigates employment discrimination.

Oscar Wilde dreamed of a world without work. In The Soul of Man Under Socialism (1891) he imagined a society liberated from drudgery by the machine: "while Humanity will be amusing itself, or enjoying cultivated leisure … or making beautiful things, or reading beautiful things, or simply contemplating the world with admiration and delight, machinery will be doing all the necessary and unpleasant work." This aesthete's Eden prompted one of his most famous observations: "Is this Utopian? A map of the world that does not include Utopia is not worth even glancing at." In Wilde's day the future of work was the first question that every aspiring utopian, from Edward Bellamy to HG Wells, needed to answer.

In a televised address to the nation from the White House, Trump emphasized there were "no Americans harmed" in the ballistic missile salvo aimed at two bases on Wednesday. While he promised to immediately impose "punishing" new economic sanctions on Tehran, Trump welcomed signs the Islamic republic "appears to be standing down" in the tit-for-tat confrontation. The comments cooled what threatened to become an uncontrolled boiling over of tensions after Trump ordered the killing last Friday of a top Iranian general, Qassem Soleimani. The president, facing both an impeachment trial in Congress and a tough re-election in November, defended his targeting of a man seen by many as Iran's second-most-influential official. Soleimani, a national hero at home, was "the world's top terrorist" and "should have been terminated long ago," Trump said.

A driverless car is displayed during a Google event in San Francisco in 2016. The Trump administration rolled out new non-binding guidelines Wednesday for regulating driverless cars and trucks -- its second move this week to advance a light-touch approach to tech regulation that contrasts with the strategy key European leaders are advocating. Transportation Secretary Elaine Chao used the big CES technology conference in Las Vegas to announce the driverless-vehicle guidelines, which her department developed in conjunction with the White House. They come just two days after the White House issued a broader set of draft principles for federal agencies' treatment of artificial intelligence, which President Donald Trump's advisers say should avoid "preemptive, burdensome or duplicative rules." The twin moves came during a week when three Cabinet secretaries and Trump's daughter Ivanka were scheduled to make appearances at the CES conference.

Artificial intelligence and machine learning (AI/ML) systems are growing exponentially around the world and is estimated to generate AU$22.17 trillion to the global economy by 2030. The Australian Government's Artificial Intelligence Technology Roadmap, developed by Data61, identified Australia's need for up to 161,000 new specialist AI workers by 2030. Stuart Ayres, NSW Minister for Jobs, Investment, Tourism and Western Sydney, said: "NeurIPS 2021 will propel Australia's research and innovative discoveries to the forefront – bringing with it opportunities for trade and investment and talent attraction as well as helping to build Sydney's brand as an intellectual capital." Dr. Terrence Sejnowski, President of the NeurIPS Foundation, agreed that it was a "significant step" bringing the conference to Australia. It will be the first time NeurIPS is held in the Asia-Pacific region, and only the third time it has been held outside North America.

On Nov. 25, an article headlined "Spot the deepfake. The editors would not have placed this piece on the front page a year ago. If they had, few would have understood what its headline meant. This technology, one of the most worrying fruits of rapid advances in artificial intelligence (AI), allows those who wield it to create audio and video representations of real people saying and doing made-up things. As this technology develops, it becomes increasingly difficult to distinguish real audio and video recordings from fraudulent misrepresentations created by manipulating real sounds and images. "In the short term, detection will be reasonably effective," says Subbarao Kambhampati, a professor of computer science at Arizona State University. "In the longer run, I think it will be impossible to distinguish between the real pictures and the fake pictures."2 The longer run may come as early as later this year, in time for the presidential election.

President Donald Trump's daughter and adviser Ivanka Trump is facing backlash over her "extreme privilege" after the first daughter was invited to participate in a keynote interview at a major technology conference in Las Vegas, Nevada. Trump participated in a keynote session at the Consumer Electronics Show (CES) on Tuesday. She discussed the "path to the future of work" with Gary Shapiro, the CEO of the Consumer Technology Association (CTA), which hosts the annual conference. The first daughter had been invited to speak at the conference given her experience helping lead the White House's approach to economic empowerment, job creation and growth. A number of prominent voices in the technology industry, however, were not convinced, with women leaders in tech branding Trump's involvement in the conference "an insult to women in technology."

Studies of networked phenomena, such as interactions in online social media, often rely on incomplete data, either because these phenomena are partially observed, or because the data is too large or expensive to acquire all at once. Analysis of incomplete data leads to skewed or misleading results. In this paper, we investigate limitations of learning to complete partially observed networks via node querying. Concretely, we study the following problem: given (i) a partially observed network, (ii) the ability to query nodes for their connections (e.g., by accessing an API), and (iii) a budget on the number of such queries, sequentially learn which nodes to query in order to maximally increase observability. We call this querying process Network Online Learning and present a family of algorithms called NOL*. These algorithms learn to choose which partially observed node to query next based on a parameterized model that is trained online through a process of exploration and exploitation. Extensive experiments on both synthetic and real world networks show that (i) it is possible to sequentially learn to choose which nodes are best to query in a network and (ii) some macroscopic properties of networks, such as the degree distribution and modular structure, impact the potential for learning and the optimal amount of random exploration.

There has recently been much work on the "wide limit" of neural networks, where Bayesian neural networks (BNNs) are shown to converge to a Gaussian process (GP) as all hidden layers are sent to infinite width. However, these results do not apply to architectures that require one or more of the hidden layers to remain narrow. In this paper, we consider the wide limit of BNNs where some hidden layers, called "bottlenecks", are held at finite width. The result is a composition of GPs that we term a "bottleneck neural network Gaussian process" (bottleneck NNGP). Although intuitive, the subtlety of the proof is in showing that the wide limit of a composition of networks is in fact the composition of the limiting GPs. We also analyze theoretically a single-bottleneck NNGP, finding that the bottleneck induces dependence between the outputs of a multi-output network that persists through extreme post-bottleneck depths, and prevents the kernel of the network from losing discriminative power at extreme post-bottleneck depths.

Innovation ecosystems can be naturally described as a collection of networked entities, such as experts, institutions, projects, technologies and products. Representing in a machine-readable form these entities and their relations is not entirely attainable, due to the existence of abstract concepts such as knowledge and due to the confidential, non-public nature of this information, but even its partial depiction is of strong interest. The representation of innovation ecosystems incarnated as knowledge graphs would enable the generation of reports with new insights, the execution of advanced data analysis tasks. An ontology to capture the essential entities and relations is presented, as well as the description of data sources, which can be used to populate innovation knowledge graphs. Finally, the application case of the Universidad Politecnica de Madrid is presented, as well as an insight of future applications.