NASA's Perseverance rover has sent back astonishing video footage of its 18 February landing on Mars. These videos give us the most intimate look ever at the process of setting a spacecraft down on the Martian surface. During the landing, five cameras took videos: two on the back of the capsule holding the rover, one on the sky crane that acted as a jet pack to lower the rover its final 2000 metres or so to the surface and two on the rover itself. The videos show the parachute opening to slow down the spacecraft, and then the heat shield dropping to the surface of Mars once Perseverance is moving slow enough not to need it anymore. "You can get a sense really of how violent that parachute deploy and inflation are," said Al Chen, a Perseverance engineer at NASA's Jet Propulsion Laboratory (JPL) in California, during a press conference.

The helicopter sent to Mars by NASA to explore the Red Planet from the sky has'phoned home' and is working great, according to the space agency. Named Ingenuity, it rode to Mars strapped to the belly of the car-sized Perseverance rover that will trundle along the Jezero crater in search of ancient alien life. NASA mission control in Southern California received the first status report from Ingenuity late on Friday via the space-based Mars Reconnaissance Orbiter. Ingenuity will remain attached the belly of Perseverance for between 30 and 60 days before it detaches and makes its maiden flight - assuming it survives the brutal average -90C overnight temperatures found on the Red Planet. NASA shared an exciting image shot by the sky crane that shows Perseverance, nicknamed Perky, slung beneath and attached to mechanical bridals – moments before making landfall. The downlink confirmed that the helicopter, and an electrical box on the rover that routes and stores communications with Earth, were both performing as expected.

Congress should pass reform legislation that would establish a Technology Innovation Fund for the U.S. Postal Service (USPS) to enable robotic last-mile postal delivery, a new report states. "Of particular promise are sorting and delivery robots, which could sort mail, including into local delivery orders, deliver mail to homes, or both," according to "A New Vision for Postal Reform in the E-commerce Age," a Feb. 11 report from the Information Technology and Innovation Foundation (ITIF). "One could imagine a postal worker driving to particular routes with a fleet of 10 or so robots, letting each one off to'walk' a particular mail route, and then picking them back up at the end of the route." This funding would help support innovation at USPS, the report states, likening the approach to those at the Defense Department and NASA, which get federal funding for automation and robotics research. Although robotics is not sophisticated or inexpensive enough yet to sort and deliver mail, progress is happening.

We consider the problem of the stability of saliency-based explanations of Neural Network predictions under adversarial attacks in a classification task. We empirically show that, for deterministic Neural Networks, saliency interpretations are remarkably brittle even when the attacks fail, i.e. for attacks that do not change the classification label. By leveraging recent results, we provide a theoretical explanation of this result in terms of the geometry of adversarial attacks. Based on these theoretical considerations, we suggest and demonstrate empirically that saliency explanations provided by Bayesian Neural Networks are considerably more stable under adversarial perturbations. Our results not only confirm that Bayesian Neural Networks are more robust to adversarial attacks, but also demonstrate that Bayesian methods have the potential to provide more stable and interpretable assessments of Neural Network predictions.

We consider the problem of domain generalization, in which a predictor is trained on data drawn from a family of related training domains and tested on a distinct and unseen test domain. While a variety of approaches have been proposed for this setting, it was recently shown that no existing algorithm can consistently outperform empirical risk minimization (ERM) over the training domains. To this end, in this paper we propose a novel approach for the domain generalization problem called Model-Based Domain Generalization. In our approach, we first use unlabeled data from the training domains to learn multi-modal domain transformation models that map data from one training domain to any other domain. Next, we propose a constrained optimization-based formulation for domain generalization which enforces that a trained predictor be invariant to distributional shifts under the underlying domain transformation model. Finally, we propose a novel algorithmic framework for efficiently solving this constrained optimization problem. In our experiments, we show that this approach outperforms both ERM and domain generalization algorithms on numerous well-known, challenging datasets, including WILDS, PACS, and ImageNet. In particular, our algorithms beat the current state-of-the-art methods on the very-recently-proposed WILDS benchmark by up to 20 percentage points.

When collaborating with an AI system, we need to assess when to trust its recommendations. If we mistakenly trust it in regions where it is likely to err, catastrophic failures may occur, hence the need for Bayesian approaches for probabilistic reasoning in order to determine the confidence (or epistemic uncertainty) in the probabilities in light of the training data. We propose an approach to overcome the independence assumption behind most of the approaches dealing with a large class of probabilistic reasoning that includes Bayesian networks as well as several instances of probabilistic logic. We provide an algorithm for Bayesian learning from sparse, albeit complete, observations, and for deriving inferences and their confidences keeping track of the dependencies between variables when they are manipulated within the unifying computational formalism provided by probabilistic circuits. Each leaf of such circuits is labelled with a beta-distributed random variable that provides us with an elegant framework for representing uncertain probabilities. We achieve better estimation of epistemic uncertainty than state-of-the-art approaches, including highly engineered ones, while being able to handle general circuits and with just a modest increase in the computational effort compared to using point probabilities.

From cultivation to improving harvesting quality, AI is known as one of the main elements for a surplus yield but that too for the ones who are capable enough to make use of it. Agriculture is seeing rapid adoption of Artificial Intelligence and Machine Learning, both in terms of agricultural products and in field farming techniques. Apart from that, most of the countries are looking forward to involving such techniques. In 2016, the estimated value added by the agricultural industry was estimated at just under 1% of the US GDP. The US Environmental Protection Agency, estimates that agriculture contributes roughly $330 billion in annual revenue to the economy, thus such techniques would definitely speed things up.

In the context of big data analysis, the divide-and-conquer methodology refers to a multiple-step process: first splitting a data set into several smaller ones; then analyzing each set separately; finally combining results from each analysis together. This approach is effective in handling large data sets that are unsuitable to be analyzed entirely by a single computer due to limits either from memory storage or computational time. The combined results will provide a statistical inference which is similar to the one from analyzing the entire data set. This article reviews some recently developments of divide-and-conquer methods in a variety of settings, including combining based on parametric, semiparametric and nonparametric models, online sequential updating methods, among others. Theoretical development on the efficiency of the divide-and-conquer methods is discussed. Examples of real-world data analyses are provided in various application areas.

Hazard detection is critical for enabling autonomous landing on planetary surfaces. Current state-of-the-art methods leverage traditional computer vision approaches to automate identification of safe terrain from input digital elevation models (DEMs). However, performance for these methods can degrade for input DEMs with increased sensor noise. At the same time, deep learning techniques have been developed for various applications. Nevertheless, their applicability to safety-critical space missions has been often limited due to concerns regarding their outputs' reliability. In response to this background, this paper proposes an uncertainty-aware learning-based method for hazard detection and landing site selection. The developed approach enables reliable safe landing site selection by: (i) generating a safety prediction map and its uncertainty map together via Bayesian deep learning and semantic segmentation; and (ii) using the generated uncertainty map to filter out the uncertain pixels in the prediction map so that the safe landing site selection is performed only based on the certain pixels (i.e., pixels for which the model is certain about its safety prediction). Experiments are presented with simulated data based on a Mars HiRISE digital terrain model and varying noise levels to demonstrate the performance of the proposed approach.

A research team of scientists from the Berkeley lab has used Artificial Intelligence(AI) to discover about 1200 possible gravitational lenses. According to phys.org if this count is accurate, this could double the number of existing gravitational lenses. Read more to find out what gravitational lenses are. So what exactly are these gravitational lenses? When light emitted by stars from distant galaxies pass massive objects in the universe, like a cluster of star systems or a bunch of galaxies, the light gets bent or distorted due to the incredibly powerful gravitational force.