The US Patent and Trademark Office (USPTO) has ruled that artificial intelligence systems cannot be credited as an inventor in a patent, the agency announced earlier this week. The decision came in response to two patents -- one for a food container and the other for a flashing light -- that were created by an AI system called DABUS. Among the USPTO's arguments is the fact that US patent law repeatedly refers to inventors using humanlike terms such as "whoever" and pronouns like "himself" and "herself." The group behind the applications had argued that the law's references to an inventor as an "individual" could be applied to a machine, but the USPTO said this interpretation was too broad. "Under current law, only natural persons may be named as an inventor in a patent application," the agency concluded. The patents were submitted last year by the Artificial Inventor Project.

The COVID-19 virus has inflicted significant pain on the US economy as airlines have cancelled thousands of flights, restaurants and retailers have shut down, and sporting events and concerts have gone silent. In response, congressional lawmakers joined together and threw an immediate lifeline of $2 trillion to small businesses, hospitals, and individuals to tackle the crisis. While Americans remain holed up in their homes, the US government and Congress are considering additional stimulus efforts to get the economy back on its feet. Additional assistance for families, businesses, and healthcare workers should and will, of course, be front and center, but one area of new stimulus should not be overlooked -- artificial intelligence. AI technology has already proven its ability to assist in the COVID-19 response by utilizing supercomputers to accelerate the research of treatments to COVID-19, as well as enabling grocery stores to better predict food and supply chain shortages.

Going to the movies was a thing we used to do -- and may do again in the future. However, if you're going to an AMC theater anywhere in the world, you won't find any Universal films. The theater chain declared war once NBCUniversal execs told The Wall Street Journal that they plan on continuing a premium VOD release strategy even after the coronavirus pandemic subsides and theaters reopen. The folks at Universal figure they can do that because Trolls World Tour made a reported $95 million since its straight-to-digital rental debut earlier this month. Warner Bros. has lined up a similar strategy for Scoob! next month, and AMC is clearly trying to scare off others from moving in that direction.

It'd be foolish to base any major health policy on one scientific study and it's unclear if this study played a role in the country's fiasco over testing--widely regarded as a major failure of the administration's COVID-19 response--but it's nonetheless alarming that it was repeated as fact by the very people we're trusting to lead our country through the pandemic. That said, the mixup isn't entirely Birx's fault; after all, the study was published in a journal after peer review and it wasn't marked on PubMed as withdrawn until weeks after the retraction occurred. The real problem here is that this study even had the prominence it did. As the co-founders of Retraction Watch, a blog that tracks academic retractions, wrote in a recent article for Wired, the case involving Birx "is a particularly dismaying and consequential example of what happens when no one bothers to engage in scientific fact-checking." "But," they cautioned, "it will not be the last time that something we thought we knew about the coronavirus because it was in a published paper will turn out to be wrong."

Deep neural network (DNN) is an indispensable machine learning tool for achieving human-level performance on many learning tasks. Yet, due to its black-box nature, it is inherently difficult to understand which aspects of the input data drive the decisions of the network. There are various real-world scenarios in which humans need to make actionable decisions based on the output DNNs. Such decision support systems can be found in critical domains, such as legislation, law enforcement, etc. It is important that the humans making high-level decisions can be sure that the DNN decisions are driven by combinations of data features that are appropriate in the context of the deployment of the decision support system and that the decisions made are legally or ethically defensible. Due to the incredible pace at which DNN technology is being developed, the development of new methods and studies on explaining the decision-making process of DNNs has blossomed into an active research field. A practitioner beginning to study explainable deep learning may be intimidated by the plethora of orthogonal directions the field is taking. This complexity is further exacerbated by the general confusion that exists in defining what it means to be able to explain the actions of a deep learning system and to evaluate a system's "ability to explain". To alleviate this problem, this article offers a "field guide" to deep learning explainability for those uninitiated in the field. The field guide: i) Discusses the traits of a deep learning system that researchers enhance in explainability research, ii) places explainability in the context of other related deep learning research areas, and iii) introduces three simple dimensions defining the space of foundational methods that contribute to explainable deep learning. The guide is designed as an easy-to-digest starting point for those just embarking in the field.

Semantic segmentation models based on convolutional neural networks (CNNs) have gained much attention in relation to remote sensing and have achieved remarkable performance for the extraction of buildings from high-resolution aerial images. However, the issue of limited generalization for unseen images remains. When there is a domain gap between the training and test datasets, CNN-based segmentation models trained by a training dataset fail to segment buildings for the test dataset. In this paper, we propose segmentation networks based on a domain adaptive transfer attack (DATA) scheme for building extraction from aerial images. The proposed system combines the domain transfer and adversarial attack concepts. Based on the DATA scheme, the distribution of the input images can be shifted to that of the target images while turning images into adversarial examples against a target network. Defending adversarial examples adapted to the target domain can overcome the performance degradation due to the domain gap and increase the robustness of the segmentation model. Cross-dataset experiments and the ablation study are conducted for the three different datasets: the Inria aerial image labeling dataset, the Massachusetts building dataset, and the WHU East Asia dataset. Compared to the performance of the segmentation network without the DATA scheme, the proposed method shows improvements in the overall IoU. Moreover, it is verified that the proposed method outperforms even when compared to feature adaptation (FA) and output space adaptation (OSA).

In this episode, Audrow Nash speaks with Janet Vertessi, Assistant Professor of Sociology at Princeton, on her book Seeing Like a Rover: How Robots, Teams, and Images Craft Knowledge of Mars. The book is written about her experience living and working with NASA's Mars Rover team, and includes her observations about the team's leadership and their relationship with their robot millions of miles away on Mars. She also gives some advice from her findings for teams. Janet Vertesi specializes in the sociology of science, knowledge, and technology. She has spent the past 7 years studying several NASA spacecraft teams as an ethnographer.

Berkeley Lab researchers (clockwise from top left) Kristin Persson, John Dagdelen, Gerbrand Ceder, and Amalie Trewartha led development of COVIDScholar, a text-mining tool for COVID-19-related scientific literature. A team of materials scientists at Lawrence Berkeley National Laboratory (Berkeley Lab) – scientists who normally spend their time researching things like high-performance materials for thermoelectrics or battery cathodes – have built a text-mining tool in record time to help the global scientific community synthesize the mountain of scientific literature on COVID-19 being generated every day. The tool, live at covidscholar.org, The hope is that the tool could eventually enable "automated science." "On Google and other search engines people search for what they think is relevant," said Berkeley Lab scientist Gerbrand Ceder, one of the project leads.

The COVID-19 pandemic has dramatically changed all of our lives -- from how we work, to how we teach our children, to how we grocery shop. As we yearn to return to normal, we're also called on to do what we can to protect ourselves, our loved ones and our communities. Los Alamos National Laboratory is no different. We have responsibilities to the nation and to the communities where we live, and we take them very seriously. As one of the largest employers in Northern New Mexico, we're doing what we can to answer the call to use our vast scientific and technical resources to help fight this disease, and protect our employees and the communities we call home. To slow the spread of the virus, we took early and aggressive measures to get as much of our workforce as possible offsite, working from home. More than 85 percent of the laboratory's workforce is teleworking. The remaining employees are onsite because they are needed to assure the safety and security of our facilities or to perform essential national security work. For those onsite, measures are in place to keep them as safe as possible following CDC guidelines.

Amazon Web Services (AWS), a subsidiary of Amazon (NasdaqGS:AMZN), has announced the general availability of Amazon Augmented Artificial Intelligence (A2I), described as "a fully managed service that makes it easy to add human review to machine learning predictions to improve model and application accuracy by continuously identifying and improving low confidence predictions." "Today, we're excited to help our customers remove another obstacle to building machine learning applications with the launch of Amazon A2I, which makes it significantly easier and faster to incorporate human judgment into machine learning applications in order to ensure higher quality predictions over a sustained period of time." Accenture Federal Services, a subsidiary of Accenture (NYSE:ACN), has won a $96 million contract from the U.S. Department of Veterans Affairs (VA) to help the latter to automate manual workflows and introduce applied intelligence (AI) and machine learning capabilities. Shawn Roman, a managing director at AFS who leads the company's work with VA, commented: "Accenture Federal Services is proud to help VA seamlessly transition its existing service management tools to both improve the user experience and increase the business value of solutions. Working with the Service Management Office, we will use human-centered design thinking methods to put the Veteran, clinician, and VA employees at the center of how services are designed and created."