We leverage recent breakthroughs in neural density estimation to propose a new unsupervised anomaly detection technique (ANODE). By estimating the probability density of the data in a signal region and in sidebands, and interpolating the latter into the signal region, a likelihood ratio of data vs. background can be constructed. This likelihood ratio is broadly sensitive to overdensities in the data that could be due to localized anomalies. In addition, a unique potential benefit of the ANODE method is that the background can be directly estimated using the learned densities. Finally, ANODE is robust against systematic differences between signal region and sidebands, giving it broader applicability than other methods. We demonstrate the power of this new approach using the LHC Olympics 2020 R\&D Dataset. We show how ANODE can enhance the significance of a dijet bump hunt by up to a factor of 7 with a 10\% accuracy on the background prediction. While the LHC is used as the recurring example, the methods developed here have a much broader applicability to anomaly detection in physics and beyond.

Artificial intelligence (AI) represents a powerful but double-edged sword as nations confront global warming, poverty and issues of peace and justice. An international team of scientists this week released a first-ever study of how AI can help--as well as hinder--sustainable development worldwide. Published today in Nature Communications, the analysis focuses on how AI impacts the 17 goals for sustainable development adopted by the United Nations in 2015. The study was co-authored by a diverse group of researchers led by Ricardo Vinuesa and Francesco Fuso Nerini, assistant professors at KTH Royal Institute of Technology. They were joined by Max Tegmark, professor at Massachusetts Institute of Technology (MIT) and author of the bestselling book Life 3.0, as well as Virginia Dignum, professor of AI Ethics at Umeå University, among other authors.

Skyqraft, a Swedish startup using AI and drones for electricity power-line inspection, has picked up $505,000 in early backing. Leading the round is "startup generator" and investor Antler, with participation from a number of angels including Claes Ekström and Tomas Kåberger. Founded in March 2019 and launched that September, Skyqraft provides what it calls "smart" infrastructure inspections for power-lines. It uses unmanned airplanes, combined with AI, to gather images and detect risk automatically. This is in contrast to the status quo, where power-lines are typically inspected by teams of people and helicopters, which isn't idea on a number of fronts. "Power-line inspections most importantly are not environmentally friendly, very costly and unsafe with the use of helicopters and people," Skyqraft co-founder and CMO Sakina Turabali tells TechCrunch.

Much of the research literature on landscape change has focused on aerial imagery or satellite imagery from remote sensing sources. This is understandable if one were to look at wide areas. However, ground-level photographs also have great value to demonstrate landscape change from a human eye-level perspective. Combing such ground-level photographs with fieldwork and spatial analysis provides the possibility to assess human-environmental factors that have led to sometimes drastic change in the landscape. Yellowstone has become one of the most popular national parks in the United States.

Researchers from Facebook and MIT Labs have proposed a new methodology that uses machine learning and satellite imagery to generate street addresses in areas of the world where individual buildings don't have a unique address. The methodology divides the street addressing into two processes. The first process is segmentation. During segmentation, road pixels are identified using deep learning from 0.5 meter resolution satellite images. The second part of segmentation involves developing the road network from these identified pixels.

The widespread use of neural networks across different scientific domains often involves constraining them to satisfy certain symmetries, conservation laws, or other domain knowledge. Such constraints are often imposed as soft penalties during model training and effectively act as domain-specific regularizers of the empirical risk loss. Physics-informed neural networks is an example of this philosophy in which the outputs of deep neural networks are constrained to approximately satisfy a given set of partial differential equations. In this work we review recent advances in scientific machine learning with a specific focus on the effectiveness of physics-informed neural networks in predicting outcomes of physical systems and discovering hidden physics from noisy data. We will also identify and analyze a fundamental mode of failure of such approaches that is related to numerical stiffness leading to unbalanced back-propagated gradients during model training. To address this limitation we present a learning rate annealing algorithm that utilizes gradient statistics during model training to balance the interplay between different terms in composite loss functions. We also propose a novel neural network architecture that is more resilient to such gradient pathologies. Taken together, our developments provide new insights into the training of constrained neural networks and consistently improve the predictive accuracy of physics-informed neural networks by a factor of 50-100x across a range of problems in computational physics. All code and data accompanying this manuscript are publicly available at \url{https://github.com/PredictiveIntelligenceLab/GradientPathologiesPINNs}.

A reflex is a simple closed loop control approach which tries to minimise an error but fails to do so because it will always react too late. An adaptive algorithm can use this error to learn a forward model with the help of predictive cues. For example a driver learns to improve their steering by looking ahead to avoid steering in the last minute. In order to process complex cues such as the road ahead deep learning is a natural choice. However, this is usually only achieved indirectly by employing deep reinforcement learning having a discrete state space. Here, we show how this can be directly achieved by embedding deep learning into a closed loop system and preserving its continuous processing. We show specifically how error back-propagation can be achieved in z-space and in general how gradient based approaches can be analysed in such closed loop scenarios. The performance of this learning paradigm is demonstrated using a line-follower both in simulation and on a real robot that show very fast and continuous learning.

Despite already having a full-fledged Japanese city named after it and being in the process of building its own Nürburgring, Toyota has unveiled plans at the 2020 Consumer Electronics Show to build a 175-acre prototype "city" as a sort of real-life laboratory for future tech. The utopia will be located at the base of Mt. Fuji, run on a connected ecosystem powered by hydrogen fuel cells, and be named Woven City. Looking like something straight out of a science fiction movie--y'know, the part at the beginning when the aliens haven't landed yet--Toyota's Woven City will initially house a population of 2,000 including company employees and their families, retired people, retailers, scientific researchers, and folks from partnering companies with room to grow. The entire thing will serve as a testbed for "autonomy, robotics, personal mobility, smart homes, and AI."

Car manufacturers are always building new prototype, y'know, cars, but this is the first time we've heard of one building its own prototype city. Toyota (obviously) is planning to build a 175-acre city at the base of Japan's Mount Fuji. Construction of'Woven City', which is being designed by the same company behind 2 World Trade Center in New York, Lego House in Denmark, and Google's Mountain View and London headquarters among others, will begin in 2021. But what exactly, we hear you cry, is the point? Toyota says Woven City will be a "living laboratory" populated by more than 2,000 people, many of whom will be scientists, engineers and researchers who will use it to "test and develop technologies such as autonomy, robotics, personal mobility, smart homes and artificial intelligence in a real-world environment".

The drone attack claimed by Yemeni rebels on key Saudi Arabian oil refineries that took place on September 14, 2019 has brought the powerful technology back into the news. Unfortunately, the strikes that disrupted roughly 5% of the world's oil supply has also contributed more ammunition to the overarching negative connotations the word "drone" conjures. "Drone" is a very broad term. Colloquially, drones are usually thought of as remote-piloted flying devices used by militaries for surveillance and offensive tactics or by civilians for recreational or business purposes. Merriam-Webster defines it as "an unmanned aircraft or ship guided by remote control or onboard computers."