In Marvel's Iron Man movie series, protagonist Tony Stark relies heavily on the artificial intelligence JARVIS for his superhero needs. Not the least of JARVIS' abilities is designing and constructing Iron Man's impressive suit of armor. Accomplishing such a task would require a deep knowledge of the physical properties of metals and metallic alloys, an incredible feat given the vast number of permutations of alloy compositions. Taking us one step closer to a real-life JARVIS, researchers at A*STAR's Institute of High Performance Computing (IHPC), together with scientists in the US and Russia, have developed a machine learning model for determining the structure-property relationship in multi-principal element alloys (MPEAs). "The emergence of high-entropy alloys and, more generally, MPEAs, is a paradigm shift in conventional alloy design," said Mehdi Jafary-Zadeh, a Scientist at IHPC.

The medical speech recognition company Nuance said Monday it will begin widely selling an artificial intelligence system to automate physician note-taking. The system, built in a partnership with Microsoft, uses technology wired into the walls of the exam room to record and build a narrative of each patient encounter that is uploaded into electronic health records. Physicians can use voice commands to fill in specific fields within the health record, including the patient's list of medical problems and medication orders. Unlock this article by subscribing to STAT Plus and enjoy your first 30 days free! STAT Plus is STAT's premium subscription service for in-depth biotech, pharma, policy, and life science coverage and analysis.

Recent years have seen breakthroughs in neural network technology: computers can now beat any living person at the most complex game invented by humankind, as well as imitate human voices and faces (both real and non-existent) in a deceptively realistic manner. Is this a victory for artificial intelligence over human intelligence? And if not, what else do researchers and developers need to achieve to make the winners in the AI race the "kings of the world?" Over the last 60 years, artificial intelligence (AI) has been the subject of much discussion among researchers representing different approaches and schools of thought. One of the crucial reasons for this is that there is no unified definition of what constitutes AI, with differences persisting even now.

WASHINGTON – The Pentagon is adopting new ethical principles as it prepares to accelerate its use of artificial intelligence technology on the battlefield. The new principles call for people to "exercise appropriate levels of judgment and care" when deploying and using AI systems, such as those that scan aerial imagery to look for targets. They also say decisions made by automated systems should be "traceable" and "governable," which means "there has to be a way to disengage or deactivate" them if they are demonstrating unintended behavior, said Air Force Lt. Gen. Jack Shanahan, director of the Pentagon's Joint Artificial Intelligence Center. The Pentagon's push to speed up its AI capabilities has fueled a fight between tech companies over a $10 billion cloud computing contract known as the Joint Enterprise Defense Infrastructure, or JEDI. Microsoft won the contract in October but hasn't been able to get started on the 10-year project because Amazon sued the Pentagon, arguing that President Donald Trump's antipathy toward Amazon and its CEO Jeff Bezos hurt the company's chances at winning the bid.

"The Five" discussed the media reaction to reports on Russia's involvement or prospective involvement in the 2020 presidential election Monday, with particular focus on cable news channels CNN and MSNBC. "In terms of these talking heads on TV, the makeup-wearing misery mongers, you're never, ever, ever going to hear them apologize for getting it wrong literally for the last four years," Fox Business Network's Dagen McDowell said. "Because in their in their arrogance and insecurity, they'll never be able to admit that they are tools for Putin and also fools." A U.S. intelligence official told Fox News Sunday that contrary to numerous recent media reports, there is no evidence to suggest that Russia is making a specific "play" to boost President Trump's reelection bid. The official added that top election security official Shelby Pierson, who briefed Congress on Russian election interference efforts earlier this month, may have overstated intelligence regarding the issue.

You might be wondering what exactly Braze.com is and why it grabs their geographic, associations and more. I found out Friday that after I composed a message on the PayPal-owned digital payments app Venmo to pay my personal trainer Jarek, Venmo passed on my geographic locations and associations (including Jarek) to Braze, which calls itself a "customer engagement" company. Just last month, the Norwegian Consumer Council issued a blistering report showing what happened to users of the dating sites OKCupid and Grindr in the background, after people revealed all about their interests. OkCupid "shared highly personal data about sexuality, drug use, political views, and more," with Braze, according to the report. Grindr, a popular dating and social app used by gay and bisexual men, sent data to Braze about the "relationship type" men were seeking on the app, per the report.

This paper considers if a reward function learned via inverse reinforcement from a human expert can be used as a feedback intervention to alter future human performance as desired (i.e., human to human apprenticeship learning). To learn reward functions two new algorithms are developed: a kernel-based inverse reinforcement learning algorithm and a Monte Carlo reinforcement learning algorithm. The algorithms are benchmarked against well-known alternatives within their respective corpus and are shown to outperform in terms of efficiency and optimality. To test the feedback intervention two randomized experiments are performed with 3,256 human participants. The experimental results demonstrate with significance that the rewards learned from "expert" individuals are effective as feedback interventions. In addition to the algorithmic contributions and successful experiments, the paper also describes three reward function modifications to improve reward function feedback interventions for humans.

While robot learning has demonstrated promising results for enabling robots to automatically acquire new skills, a critical challenge in deploying learning-based systems is scale: acquiring enough data for the robot to effectively generalize broadly. Imitation learning, in particular, has remained a stable and powerful approach for robot learning, but critically relies on expert operators for data collection. In this work, we target this challenge, aiming to build an imitation learning system that can continuously improve through autonomous data collection, while simultaneously avoiding the explicit use of reinforcement learning, to maintain the stability, simplicity, and scalability of supervised imitation. To accomplish this, we cast the problem of imitation with autonomous improvement into a multi-task setting. We utilize the insight that, in a multi-task setting, a failed attempt at one task might represent a successful attempt at another task. This allows us to leverage the robot's own trials as demonstrations for tasks other than the one that the robot actually attempted. Using an initial dataset of multi-task demonstration data, the robot autonomously collects trials which are only sparsely labeled with a binary indication of whether the trial accomplished any useful task or not. We then embed the trials into a learned latent space of tasks, trained using only the initial demonstration dataset, to draw similarities between various trials, enabling the robot to achieve one-shot generalization to new tasks. In contrast to prior imitation learning approaches, our method can autonomously collect data with sparse supervision for continuous improvement, and in contrast to reinforcement learning algorithms, our method can effectively improve from sparse, task-agnostic reward signals.

Topic models are widely used to discover the latent representation of a set of documents. The two canonical models are latent Dirichlet allocation, and Gaussian latent Dirichlet allocation, where the former uses multinomial distributions over words, and the latter uses multivariate Gaussian distributions over pre-trained word embedding vectors as the latent topic representations, respectively. Compared with latent Dirichlet allocation, Gaussian latent Dirichlet allocation is limited in the sense that it does not capture the polysemy of a word such as ``bank.'' In this paper, we show that Gaussian latent Dirichlet allocation could recover the ability to capture polysemy by introducing a hierarchical structure in the set of topics that the model can use to represent a given document. Our Gaussian hierarchical latent Dirichlet allocation significantly improves polysemy detection compared with Gaussian-based models and provides more parsimonious topic representations compared with hierarchical latent Dirichlet allocation. Our extensive quantitative experiments show that our model also achieves better topic coherence and held-out document predictive accuracy over a wide range of corpus and word embedding vectors.

In recent years, different types of adversarial examples from different fields have emerged endlessly, including purely natural ones without perturbations. A variety of defenses are proposed and then broken quickly. Two fundamental questions need to be asked: What's the reason for the existence of adversarial examples and are adversarial examples unsolvable? In this paper, we will show the reason for the existence of adversarial examples is there are non-isomorphic natural explanations that can all explain data set. Specifically, for two natural explanations of being true and provable, G\"odel's sentence is an adversarial example but ineliminable. It can't be solved by the re-accumulation of data set or the re-improvement of learning algorithm. Finally, from the perspective of computability, we will prove the incomputability for adversarial examples, which are unrecognizable.