D. Tyler McQuade, Ph.D, professor in the Department of Chemical and Life Science Engineering at Virginia Commonwealth University College of Engineering, is principal investigator of a multi-university project seeking to use artificial intelligence to help scientists come up with the perfect molecule for everything from a better shampoo to coatings on advanced microchips. The project is one of the first in the U.S. to be selected for $994,433 in funding as part of a new pilot project of the National Science Foundation (NSF) called the Convergence Accelerator (C-Accel). McQuade and his collaborators will pitch their prototype in March 2020 in a bid for additional funding of up to $5 million over five years. Adam Luxon, a Ph.D. student in the Department of Chemical and Life Science Engineering who has been involved from the beginning, explained it this way: "We want to essentially make the Alexa of chemistry." Just as Amazon, Google and Netflix use data algorithms to suggest customized predictions, the team plans to build a platform and open knowledge network that can combine and help users make sense of molecular sciences data pulled from a wide range of sources including academia, industry and government.

On August 29, 2019 I put on a workshop on Emerging Memories and Artificial Intelligence at Stanford University put on by the Stanford Center for Magnetic Nanotechnology and Coughlin Associates. We had several interesting speakers talking about various types of artificial intelligence and the role that new non-volatile memories will play in both training AI models and implementing them in the field using inference engines. This piece will talk about some of the material presented at this workshop. Dr. Shan Wang, co-organizer of the event gave a introduction, talking about emerging non-volatile memories and in particular on Magnetic Random Access Memory (MRAM). He spoke about how various new memories work--in particular Resistive RAM (RRAM), Phase Change Memory (PCM), MRAM and Ferrroelectic RAM (FRAM).

On Friday, September 6, 2019, Secretary of Energy Rick Perry announced the formal establishment of DOE's Artificial Intelligence and Technology Office. Chief Technology Officer of the United States, Michael Kratsios: "Under Secretary Perry's leadership, the Department of Energy is critical to our Nation's success in the development and application of artificial intelligence. I commend the Secretary for establishing the Artificial Intelligence and Technology Office to coordinate DOE's vast AI efforts and carry out the mission of the Trump Administration's national strategy for American leadership in AI." Representative Bill Foster (D-IL): "Developments in artificial intelligence are at the forefront of technological innovations that are changing our country and our world. The Department of Energy and our world-class national laboratories are leading the way, including in the district I represent, where Argonne National Laboratory is building the world's most advanced supercomputer to power the use of AI for everything from responding to climate change, to fighting cancer, and ensuring our nation's security. DOE's new Artificial Intelligence and Technology Office will play a crucial role in coordinating the work being done in AI and provide support for the researchers across DOE who are on the frontlines of this transformative work. I applaud Secretary Perry for continuing DOE's long history of scientific leadership by establishing this new office, and I look forward to working in Congress to support AITO's work, as well as the continued technological leadership of our national labs."

New deep learning models predict the interactions between atoms in organic molecules. These models, which were generated using supercomputers at the San Diego Supercomputer Center and the Los Alamos National Laboratory, help computational biologists and drug development researchers better understand and treat disease. According to a release issued earlier this month by the Los Alamos National Laboratory (LANL), researchers have developed a machine learning approach called transfer learning that lets them model novel materials by learning from data collected about millions of other compounds. The new approach can be applied to new molecules in milliseconds, enabling research into a far greater number of compounds over much longer timescales. The new technique, called ANI-1ccx potential, promises to advance the capabilities of researchers in many fields and improve the accuracy of machine learning-based potentials in future studies of metal alloys and detonation physics.

With over 1 billion active iOS powered device users and 2 billion active Android-powered device users, the custom mobile app development sector is providing the most profitable and captivating markets to develop and sell the most advanced digital solutions to the users all across the globe. There are about 4 million unique mobile applications on this operating system, most of which have similar functions. Machine learning is a crucial and integral part of Artificial Intelligence which are widely used in software development services. Also, it had an impact on mobile applications development and has become a potential market for implementing software that can adapt to user behavior. Now these days, everyone wants a customized user experience according to their specific needs.

Artificial intelligence (AI) has arguably become one of the most overhyped and over-marketed terms of the last couple of years. Yet for all its flash, AI is capable of strengthening our energy security by increasing power grid resilience and reducing the likelihood blackouts caused by energy surges and shortages. Power grids are the arteries through which the modern economy pulses. They are responsible for distributing and transmitting electricity from power plants to substations and, finally, to consumers. Given our acute dependency on an uninterrupted supply of power, it is hardly surprising that power grids are among the most strategically important pieces of infrastructure for economic and national security alike.

In the 1970s, the U.S. Department of Energy poured money into making practical a miraculous technology: the ability to convert sunlight into electricity. Solar energy was a pipe dream, far too expensive and unreliable to be considered a practical power source. But yesterday's moon shot is today's reality. The expense of solar power has fallen more quickly than expected, with installations costing about 80% less today than a decade ago. Alternative energy (like wind and solar) is now often cheaper than conventional energy (like coal and gas).

Traditionally, off-policy learning algorithms (such as Q-learning) and exploration schemes have been derived separately. Often times, the exploration-exploitation dilemma being addressed through heuristics. In this article we show that both the learning equations and the exploration-exploitation strategy can be derived in tandem as the solution to a unique and well-posed optimization problem whose minimization leads to the optimal value function. We present a new algorithm following this idea. The algorithm is of the gradient type (and therefore has good convergence properties even when used in conjunction with function approximators such as neural networks); it is off-policy; and it specifies both the update equations and the strategy to address the exploration-exploitation dilemma. To the best of our knowledge, this is the first algorithm that has these properties.

A new era in radioastronomy will begin with the upcoming large-scale surveys planned at the Australian Square Kilometre Array Pathfinder (ASKAP). ASKAP started its Early Science program in October 2017 and several target fields were observed during the array commissioning phase. The SCORPIO field was the first observed in the Galactic Plane in Band 1 (792-1032 MHz) using 15 commissioned antennas. The achieved sensitivity and large field of view already allow to discover new sources and survey thousands of existing ones with improved precision with respect to previous surveys. Data analysis is currently ongoing to deliver the first source catalogue. Given the increased scale of the data, source extraction and characterization, even in this Early Science phase, have to be carried out in a mostly automated way. This process presents significant challenges due to the presence of extended objects and diffuse emission close to the Galactic Plane. In this context we have extended and optimized a novel source finding tool, named CAESAR , to allow extraction of both compact and extended sources from radio maps. A number of developments have been done driven by the analysis of the SCORPIO map and in view of the future ASKAP Galactic Plane survey. The main goals are the improvement of algorithm performances and scalability as well as of software maintainability and usability within the radio community. In this paper we present the current status of CAESAR and report a first systematic characterization of its performance for both compact and extended sources using simulated maps. Future prospects are discussed in light of the obtained results.

Data fields sampled on irregularly spaced points arise in many applications in the sciences and engineering. For regular grids, Convolutional Neural Networks (CNNs) have been successfully used to gaining benefits from weight sharing and invariances. We generalize CNNs by introducing methods for data on unstructured point clouds based on Generalized Moving Least Squares (GMLS). GMLS is a non-parametric technique for estimating linear bounded functionals from scattered data, and has recently been used in the literature for solving partial differential equations. By parameterizing the GMLS estimator, we obtain learning methods for operators with unstructured stencils. In GMLS-Nets the necessary calculations are local, readily parallelizable, and the estimator is supported by a rigorous approximation theory. We show how the framework may be used for unstructured physical data sets to perform functional regression to identify associated differential operators and to regress quantities of interest. The results suggest the architectures to be an attractive foundation for data-driven model development in scientific machine learning applications.