In one of the most profound examples, the first image of a black hole was captured by applying a machine learning algorithm to petabytes of data collected from eight telescopes [1]. Since planetary science's last decadal survey, the use of machine learning has increased in each division of NASA's Science Mission Directorate (SMD). However, even though the number of planetary science publications involving machine learning has grown exponentially over the last ten years, it lags in both percent share and growth rate compared to heliophysics, astrophysics, and Earth science (Figure 1). In this white paper, we assert that planetary science, similar to related disciplines, has the opportunity to leverage machine learning methods for scientific advancement in our field.

In this paper we introduce Q-Rock, a development cycle for the automated self-exploration and qualification of robotic behaviors. With Q-Rock, we suggest a novel, integrative approach to automate robot development processes. Q-Rock combines several machine learning and reasoning techniques to deal with the increasing complexity in the design of robotic systems. The Q-Rock development cycle consists of three complementary processes: (1) automated exploration of capabilities that a given robotic hardware provides, (2) classification and semantic annotation of these capabilities to generate more complex behaviors, and (3) mapping between application requirements and available behaviors. These processes are based on a graph-based representation of a robot's structure, including hardware and software components. A graph-database serves as central, scalable knowledge base to enable collaboration with robot designers including mechanical and electrical engineers, software developers and machine learning experts. In this paper we formalize Q-Rock's integrative development cycle and highlight its benefits with a proof-of-concept implementation and a use case demonstration.

In a report published today by the National Institutes of Science and Technology (NIST), a physical sciences laboratory and non-regulatory agency of the U.S. Department of Commerce, researchers attempted to evaluate the performance of facial recognition algorithms on faces partially covered by protective masks. They report that the 89 commercial facial recognition algorithms from Panasonic, Canon, Tencent, and others they tested had error rates between 5% and 50% in matching digitally applied masks with photos of the same person without a mask. "With the arrival of the pandemic, we need to understand how face recognition technology deals with masked faces," Mei Ngan, a NIST computer scientist and a coauthor of the report, said in a statement. "We have begun by focusing on how an algorithm developed before the pandemic might be affected by subjects wearing face masks. Later this summer, we plan to test the accuracy of algorithms that were intentionally developed with masked faces in mind."

Strong engagement operates with the understanding that participants have a mandate to drive change and an influence on the policymaking and the decisions ultimately made. Participants may have different roles, including informing, consulting, involving, collaborating or empowering. It's important designers are clear about the role they want participants to play and the level of influence they'll have. Without that clarity, trust will be lost. Engagement drives a strong "before and after," where discussions are linked to outcomes requested by governments, businesses and the people.

Protesters in Austin, Texas; Tempe, Ariz.; and Portland, Ore., took the streets in their respective cities to march in step with the Black Lies Matter movement. In some instances, police and federal agents clashed with protesters, but in one demonstration, protesters paid their respects to one of their own in Texas. Coronavirus face masks can confuse facial recognition technology, government researchers announced Monday after a preliminary study on the issue. Facial recognition algorithms developed before the outbreak struggle to identify people wearing masks or face coverings, according to a new study from the U.S. Commerce Department's National Institute of Standards and Technology (NIST). The NIST looked at 89 commercial algorithms and found that even the best had error rates of around 5 percent when trying to match a masked individual's face with their unmasked appearance – up from a normal rate of 0.3 percent when trying to match two photos of the same person.

Repeated activity wears on soft robotic actuators, but these machines' moving parts need to be reliable and easily fixed. Now a team of researchers has a biosynthetic polymer, patterned after squid ring teeth, that is self-healing and biodegradable, creating a material not only good for actuators, but also for hazmat suits and other applications where tiny holes could cause a danger. "Current self-healing materials have shortcomings that limit their practical application, such as low healing strength and long healing times (hours)," the researchers report in today's (July 27) issue of Nature Materials. The researchers produced high-strength synthetic proteins that mimic those found in nature. Like the creatures they are patterned on, the proteins can self-heal both minute and visible damage.

With a $740.5 billion budget for national security, the United States continues to be the leading country in terms of combat power. The US Air Force has just signed four contractors to build an unmanned combat aircraft with artificial intelligence (AI) for as much as $400 million. With the initiative to create a low-cost combat aircraft, with modular payloads for a multitude of air and ground-attack capabilities, the Skyborg Vanguard program was created. Skyborg is an autonomy-focused aircraft that will enable the Air Force to operate unmanned teamed aircrafts at a sustainable low cost. The program is undergoing prototyping whereby they are developing an autonomous aircraft which is equipped with unmanned system technologies to support a range of Air Force missions.

Artificial intelligence is proving to be an integral part of the fight against the ongoing COVID-19 crisis and may also aid in battling future pandemics, according to researchers from a high-profile AI commission. "The early months of the pandemic response suggest that technology -- some of it underpinned by artificial intelligence -- offers powerful potential for detecting and containing the virus, driving biomedical innovation -- including for vaccines and therapeutics -- and improving response and recovery," said a new white paper by the National Security Commission on Artificial Intelligence. The panel -- which was established by the fiscal year 2019 National Defense Authorization Act -- was tasked by Congress to research ways to advance the development of AI for defense purposes. In the report titled, "The Role of AI Technology in Pandemic Response and Preparedness: Recommended Investments and Initiatives," the authors called for a number of focused spending efforts and projects that could help the United States capitalize on AI capabilities during pandemics, as well as maintain military readiness. Already, the Pentagon has a number of initiatives examining such tools.

A leading global corporate, fiduciary and fund administrator has selected Predict360's AI-powered Regulatory Change Management solution to track, evaluate and manage regulatory requirements and changes, 360factors announced today. With its unique mapping abilities and Artificial Intelligence (AI) with Natural Language Processing (NLP) technology, Predict360 streamlines the process of regulatory change management (RCM) by first documenting regulatory changes and updates via external sources such as news feeds, then providing a mechanism to document applicability. An audit trail that associates users with change assessments provides an accountability layer to the entire organization. If a regulatory change is determined to be applicable to the organization, the solution will automatically trigger a regulatory change workflow process that enables stakeholders to manage project plans across business units and product lines along with related artifacts such as, documents, policies and procedures, assessment checklists and more. The solution provides proactive notifications to all relevant parties across the organization that are associated with that regulation and/or change, including due dates, effective dates, action items, and supporting documentation.

In multimedia forensics, learning-based methods provide state-of-the-art performance in determining origin and authenticity of images and videos. However, most existing methods are challenged by out-of-distribution data, i.e., with characteristics that are not covered in the training set. This makes it difficult to know when to trust a model, particularly for practitioners with limited technical background. In this work, we make a first step toward redesigning forensic algorithms with a strong focus on reliability. To this end, we propose to use Bayesian neural networks (BNN), which combine the power of deep neural networks with the rigorous probabilistic formulation of a Bayesian framework. Instead of providing a point estimate like standard neural networks, BNNs provide distributions that express both the estimate and also an uncertainty range. We demonstrate the usefulness of this framework on a classical forensic task: resampling detection. The BNN yields state-of-the-art detection performance, plus excellent capabilities for detecting out-of-distribution samples. This is demonstrated for three pathologic issues in resampling detection, namely unseen resampling factors, unseen JPEG compression, and unseen resampling algorithms. We hope that this proposal spurs further research toward reliability in multimedia forensics.