The first phase of the UNOSAT Challenge has just ended. The UNOSAT Challenge is the important Phi-Unet (ESA) contest for UNOSAT (United Nations) in partnership with ESA, RUS COPERNICUS, UNOSAT and with the technical support of CERN openlab. The aim of the contest is to put artificial intelligence and Earth Observation data at the service of a humanitarian cause: support the Iraqi government in planning reconstruction activities. In the first phase, candidates were asked to create an artificial intelligence model to identify urban areas in some Iraqi territories, working on data provided by ESA and the German Space Agency (DLR). At the end of this phase, 5 teams were selected.

AI and machine learning are the kind of buzzwords that generate a lot of interest; hence, they get thrown around all the time. But what do they actually mean? And are they as instrumental to the future of cybersecurity as many believe? When a large set of data is involved, having to analyze it all by hand seems like a nightmare. It's the kind of work that one would describe as boring and tedious. Not to mention the fact it would take a lot of staring at the screen to find what you've set out to discover.

An Aussie entrepreneur is copping flack online for his contentious and, frankly, dystopian startup designed to identify people and source information about them, from a single image. According to The New York Times, the technology has already been provided to more than 600 law enforcement agencies, including local police in Florida, the FBI and the Department of Homeland Security. Founded by Hoan Ton-That, Clearview AI is a secretive Silicon Valley startup that has been reportedly operating in stealth mode for some time. It's facial recognition app allows users to take a picture of a person and upload it, to access public photos of that person, and the sites on which they appear (think Facebook and YouTube). It has a database of about 3 billion images.

Finding the intersection point between the worlds of digital intelligence and human empathy is ... [ ] surely the biggest challenge on the automated road ahead. The world is obviously going through some changeable times. The United Kingdom is about to ride through the (many would argue) uncertain period of post-Brexit'independence' and global geopolitical swings continue to have an impact upon international trade and investment. The technology industry thinks it can help, well, when doesn't it? In particular, the tech business is keen to extol the virtues of Artificial Intelligence (AI), Machine Learning (ML) and Robotic Process Automation (RPA) as key tools to help manage the things that humans shouldn't be troubling themselves with.

WASHINGTON – The House Ethics Committee is warning lawmakers not to share doctored images or videos that could "erode public trust, effect public discourse, or sway an election," guidance that comes during a proliferation of online misinformation in the run-up to the 2020 elections. In a memo sent to House members Tuesday, the committee said lawmakers or staffers could be found in violation of House ethics rules and subject to disciplinary proceedings for posting content intended to mislead the public. "Members have a duty, and a First Amendment right, to contribute to the public discourse," the authors of the memo wrote. "However, manipulation of images and videos that are intended to mislead the public can harm that discourse and reflect discreditably on the House." It's the first time the committee has admonished members of Congress on the use of fake images and audio on social media, though the new guidelines may be difficult to enforce because of a loophole allowing fake images when used for satire or parody.

The Interior Department is grounding its fleet of drones -- including Chinese-made models such as this specialized "government edition" Mavic Pro, made by DJI. The Interior Department is grounding its fleet of drones -- including Chinese-made models such as this specialized "government edition" Mavic Pro, made by DJI. The Interior Department has grounded its fleet of more than 800 drones, citing potential cybersecurity risks and the need to support U.S. drone production – suggesting the move is aimed at least in part at China, a leading drone producer. Interior Secretary David Bernhardt signed an order on Wednesday grounding the drones, formalizing a "pause" he ordered nearly three months ago. "We've had only 12 drone flights since that time for emergency operations related to fires and floods," DOI Spokesperson Carol Danko said. In 2018, the Interior Department reported making 10,432 drone flights -- part of a broad expansion of its use of the aircraft under President Trump.

Joseph Bakarji 1, Daniel M. Tartakovsky 1 Department of Energy Resources Engineering, Stanford University, 367 Panama Mall, Stanford, 94305 CA, USAAbstract A general method for learning probability density function (PDF) equations based on Monte Carlo simulations of random fields is proposed. Sparse linear regression is used to discover the relevant terms of a partial differential equation of the distribution. The various properties of PDF equations, like smoothness and conservation, makes them very well adapted to equation learning methods. The results show a promising direction for data-driven discovery of coarse-grained equations in general. Introduction Probabilistic models have proven to be essential in various fields of science and technology for optimizing predictability under epistemic and model uncertainty.

Gaussian process (GP) priors are non-parametric generative models with appealing modelling properties for Bayesian inference: they can model non-linear relationships through noisy observations, have closed-form expressions for training and inference, and are governed by interpretable hyperparameters. However, GP models rely on Gaussianity, an assumption that does not hold in several real-world scenarios, e.g., when observations are bounded or have extreme-value dependencies, a natural phenomenon in physics, finance and social sciences. Although beyond-Gaussian stochastic processes have caught the attention of the GP community, a principled definition and rigorous treatment is still lacking. In this regard, we propose a methodology to construct stochastic processes, which include GPs, warped GPs, Student-t processes and several others under a single unified approach. We also provide formulas and algorithms for training and inference of the proposed models in the regression problem. Our approach is inspired by layers-based models, where each proposed layer changes a specific property over the generated stochastic process. That, in turn, allows us to push-forward a standard Gaussian white noise prior towards other more expressive stochastic processes, for which marginals and copulas need not be Gaussian, while retaining the appealing properties of GPs. We validate the proposed model through experiments with real-world data.

The development of assessment methods for the performance of Automated Vehicles (AVs) is essential to enable and speed up the deployment of automated driving technologies, due to the complex operational domain of AVs. As traditional methods for assessing vehicles are not applicable for AVs, other approaches have been proposed. Among these, real-world scenario-based assessment is widely supported by many players in the automotive field. In this approach, test cases are derived from real-world scenarios that are obtained from driving data. To minimize any ambiguity regarding these test cases and scenarios, a clear definition of the notion of scenario is required. In this paper, we propose a more concrete definition of scenario, compared to what is known to the authors from the literature. This is achieved by proposing an ontology in which the quantitative building blocks of a scenario are defined. An example illustrates that the presented ontology is applicable for scenario-based assessment of AVs.

Active Learning (AL) is a human-in-the-loop Machine Learning paradigm favored for its ability to learn with fewer labeled instances, but the model's states and progress remain opaque to the annotators. Meanwhile, many recognize the benefits of model transparency for people interacting with ML models, as reflected by the surge of explainable AI (XAI) as a research field. However, explaining an evolving model introduces many open questions regarding its impact on the annotation quality and the annotator's experience. In this paper, we propose a novel paradigm of explainable active learning (XAL), by explaining the learning algorithm's prediction for the instance it wants to learn from and soliciting feedback from the annotator. We conduct an empirical study comparing the model learning outcome, human feedback content and the annotator experience with XAL, to that of traditional AL and coactive learning (providing the model's prediction without the explanation). Our study reveals benefits--supporting trust calibration and enabling additional forms of human feedback, and potential drawbacks--anchoring effect and frustration from transparent model limitations--of providing local explanations in AL. We conclude by suggesting directions for developing explanations that better support annotator experience in AL and interactive ML settings.