Intelligent robots are redefining a multitude of critical domains but are still far from being fully capable of assisting human peers in day-to-day tasks. An important requirement of collaboration is for each teammate to maintain and respect an understanding of the others' expectations of itself. Lack of which may lead to serious issues such as loose coordination between teammates, reduced situation awareness, and ultimately teaming failures. Hence, it is important for robots to behave explicably by meeting the human's expectations. One of the challenges here is that the expectations of the human are often hidden and can change dynamically as the human interacts with the robot. However, existing approaches to generating explicable plans often assume that the human's expectations are known and static. In this paper, we propose the idea of active explicable planning to relax this assumption. We apply a Bayesian approach to model and predict dynamic human belief and expectations to make explicable planning more anticipatory. We hypothesize that active explicable plans can be more efficient and explicable at the same time, when compared to explicable plans generated by the existing methods. In our experimental evaluation, we verify that our approach generates more efficient explicable plans while successfully capturing the dynamic belief change of the human teammate.

Vision transformers (ViTs) have demonstrated impressive performance on a series of computer vision tasks, yet they still suffer from adversarial examples. In this paper, we posit that adversarial attacks on transformers should be specially tailored for their architecture, jointly considering both patches and self-attention, in order to achieve high transferability. More specifically, we introduce a dual attack framework, which contains a Pay No Attention (PNA) attack and a PatchOut attack, to improve the transferability of adversarial samples across different ViTs. We show that skipping the gradients of attention during backpropagation can generate adversarial examples with high transferability. In addition, adversarial perturbations generated by optimizing randomly sampled subsets of patches at each iteration achieve higher attack success rates than attacks using all patches. We evaluate the transferability of attacks on state-of-the-art ViTs, CNNs and robustly trained CNNs. The results of these experiments demonstrate that the proposed dual attack can greatly boost transferability between ViTs and from ViTs to CNNs. In addition, the proposed method can easily be combined with existing transfer methods to boost performance.

Fox News anchor Bret Baier offers analysis on that and other breaking news stories, on'Your World'. Three of the five prominent Sunday morning newscasts avoided the explosive New York Times report about the botched U.S. drone strike the Pentagon finally admitted killed Afghan civilians rather than ISIS-K terrorists the Biden administration previously touted. During a Friday press conference, the Pentagon confirmed that the Aug. 28 drone strike was a "tragic mistake" that killed ten civilians, including seven children, which was meant to be in response to the Aug. 26 terrorist attack outside the Kabul airport that left 13 U.S. servicemen dead. This came one week after the Times published a stunning visual investigation that came to the same conclusion. The Biden administration had announced that "two high profile" ISIS-K fighters who were dubbed as "planners and facilitators" of the suicide bombing were killed in the strike.

Each year, national and local governments determine the relative priorities of services to allocate funding. How would AI spend the cash? What makes more sense for a vibrant society -- spending on economic development or growth, spending on education, international development, social care, libraries? What would ideal balance look like? If this question sounds familiar, it's not the first time we've tried to apply artificial intelligence (AI) to making this decision.

Head of the United States Central Command Gen. Kenneth McKenzie announced Friday that it is unlikely any ISIS-K members were killed in a Kabul drone strike on August 29, which led to the deaths of multiple civilian casualties. "We now assess that it is unlikely that the vehicle and those who died were associated with ISIS-K or a direct threat to US forces," McKenzie said of the airstrike at a briefing. The drone strike, which was intended to target ISIS-K operatives, resulted in the deaths of an aid worker and up to nine of his family members, including seven children.

A US Central Command inquiry found the aid worker's car had been seen at a compound associated with IS-K, and its movements aligned with other intelligence about the terror group's plans for an attack on Kabul airport.

The International Telecommunication Union (ITU) recently connected with pioneering artificial intelligence (AI) expert Kai-Fu Lee, former president of Google China, CEO of Sinovation Ventures, and co-author of the forthcoming book "AI 2041: Ten visions for our future." Here, Lee shares insights on how much – and how deeply – AI could shape our world in the decades to come. Kai-Fu Lee: Classic childhood companions like Barbie or GI Joe – once inanimate objects - will come to life on mobile phone screens, or through virtual reality (VR) or augmented reality (AR) glasses. This can help kids learn things in a fun way – like multiplication or division – before going into school. Or recast problems: math equations could become basketball games.

U.S. household debt is almost $15 trillion and two-thirds of that is mortgage debt. Pandemic-related unemployment, high inflation and natural disasters increase lenders' credit risk as the deadline to end mortgage forbearance nears. With two million household mortgages in forbearance resulting from pandemic relief, plus a rising number of originations, it is increasingly important to deploy technology that monitors credit risk and predicts delinquency in advance. The Federal Reserve Bank of New York's (NY Fed) Center for Microeconomic Data released its Q2 2021 Quarterly Report on Household Debt and Credit on August 4th, 2021. Delinquency numbers are the lowest they've been since 2006, but household debt has risen to an all-time high.

A policy regulating the use of artificial intelligence (AI) in healthcare has been launched in Dubai. AI in healthcare includes the use of robots to perform surgery faster and assist the surgeon more efficiently; and data-based analysis of health information. Dr Mohammad Al Redha, director of Health Informatics and Smart Health Department at the Dubai Health Authority (DHA), said AI refers to systems or devices that simulate human intelligence to perform tasks with the use of data. The policy aims to determine regulatory requirements for the provision of AI solutions healthcare. It lays down the ethical requirements and defines the main roles and responsibilities of stakeholders.

As a data scientist, you have several basic tools at your disposal, which you can also apply in combination to a data set. More and more complex dependencies are formed. This makes it all the more difficult to recognize these similar properties and to assign the data to so-called clusters in a way that can be evaluated. You have certainly heard of these algorithms and maybe used one or the other, but do you really know what clustering algorithms are? So let's first clarify what these algorithms are in the first place.