Tables provide valuable knowledge that can be used to verify textual statements. While a number of works have considered table-based fact verification, direct alignments of tabular data with tokens in textual statements are rarely available. Moreover, training a generalized fact verification model requires abundant labeled training data. In this paper, we propose a novel system to address these problems. Inspired by counterfactual causality, our system identifies token-level salience in the statement with probing-based salience estimation. Salience estimation allows enhanced learning of fact verification from two perspectives. From one perspective, our system conducts masked salient token prediction to enhance the model for alignment and reasoning between the table and the statement. From the other perspective, our system applies salience-aware data augmentation to generate a more diverse set of training instances by replacing non-salient terms. Experimental results on TabFact show the effective improvement by the proposed salience-aware learning techniques, leading to the new SOTA performance on the benchmark. Our code is publicly available at https://github.com/luka-group/Salience-aware-Learning .

Internet of Things (IoT) based applications face an increasing number of potential security risks, which need to be systematically assessed and addressed. Expert-based manual assessment of IoT security is a predominant approach, which is usually inefficient. To address this problem, we propose an automated security assessment framework for IoT networks. Our framework first leverages machine learning and natural language processing to analyze vulnerability descriptions for predicting vulnerability metrics. The predicted metrics are then input into a two-layered graphical security model, which consists of an attack graph at the upper layer to present the network connectivity and an attack tree for each node in the network at the bottom layer to depict the vulnerability information. This security model automatically assesses the security of the IoT network by capturing potential attack paths. We evaluate the viability of our approach using a proof-of-concept smart building system model which contains a variety of real-world IoT devices and potential vulnerabilities. Our evaluation of the proposed framework demonstrates its effectiveness in terms of automatically predicting the vulnerability metrics of new vulnerabilities with more than 90% accuracy, on average, and identifying the most vulnerable attack paths within an IoT network. The produced assessment results can serve as a guideline for cybersecurity professionals to take further actions and mitigate risks in a timely manner.

In the last few years, open-domain question answering (ODQA) has advanced rapidly due to the development of deep learning techniques and the availability of large-scale QA datasets. However, the current datasets are essentially designed for synchronic document collections (e.g., Wikipedia). Temporal news collections such as long-term news archives spanning several decades, are rarely used in training the models despite they are quite valuable for our society. In order to foster the research in the field of ODQA on such historical collections, we present ArchivalQA, a large question answering dataset consisting of 1,067,056 question-answer pairs which is designed for temporal news QA. In addition, we create four subparts of our dataset based on the question difficulty levels and the containment of temporal expressions, which we believe could be useful for training or testing ODQA systems characterized by different strengths and abilities. The novel QA dataset-constructing framework that we introduce can be also applied to create datasets over other types of collections.

Constructing adversarial perturbations for deep neural networks is an important direction of research. Crafting image-dependent adversarial perturbations using white-box feedback has hitherto been the norm for such adversarial attacks. However, black-box attacks are much more practical for real-world applications. Universal perturbations applicable across multiple images are gaining popularity due to their innate generalizability. There have also been efforts to restrict the perturbations to a few pixels in the image. This helps to retain visual similarity with the original images making such attacks hard to detect. This paper marks an important step which combines all these directions of research. We propose the DEceit algorithm for constructing effective universal pixel-restricted perturbations using only black-box feedback from the target network. We conduct empirical investigations using the ImageNet validation set on the state-of-the-art deep neural classifiers by varying the number of pixels to be perturbed from a meagre 10 pixels to as high as all pixels in the image. We find that perturbing only about 10% of the pixels in an image using DEceit achieves a commendable and highly transferable Fooling Rate while retaining the visual quality. We further demonstrate that DEceit can be successfully applied to image dependent attacks as well. In both sets of experiments, we outperformed several state-of-the-art methods.

Swarms of drones have gotten a lot of time in the spotlight lately, mostly for their use in potential military operations. The U.S. military is testing out swarm operations in simulations, while the British Army is using live drones operating in swarms during actual training operations. Other militaries are also interested in deploying swarms. One of the biggest advantages a swarm of drones has when performing military operations is its resiliency. If a swarm enters combat and several individual drones get shot down or otherwise incapacitated, it really doesn't reduce the combat effectiveness of the swarm, nor the tactics that it uses.

The Biden administration's focus on science will include a strong emphasis on artificial intelligence. The Commerce Department, National AI Initiative Office and White House Office of Science and Technology Policy are forming a National Artificial Intelligence Advisory Committee (NAIAC) to advise the President and federal officials on AI-related issues. NAIAC will provide guidance on several AI concerns, including "competitiveness," employment, scientific progress, the viability of national strategy and future initiative revisions. The committee will also address ethical issues ranging from workforce equity to accountability and algorithmic bias. Members will come from a "broad and interdisciplinary" pool including academics, companies, non-profits and federal labs.

The US Department of Defense has released footage of an uncrewed ship firing a large missile, in a demonstration of its Ghost Fleet Overlord programme, an initiative to develop robot vessels that can operate alongside crewed warships. Previous Ghost Fleet operations have focused on endurance missions without human assistance, including the first uncrewed transit of the Panama Canal, but the firing is the first indication that the vessels will be armed. The SM-6 weapon used in the demonstration is a 1500-kilogram missile travelling at Mach 3.5 with …

A new study warns that the American AI industry is highly concentrated in the San Francisco Bay Area and that this could prove to be a weakness in the long run. The Bay leads all other regions of the country in AI research and investment activity, accounting for about one-quarter of AI conference papers, patents, and companies in the US. Bay Area metro areas see levels of AI activity four times higher than other top cities for AI development. "When you have a high percentage of all AI activity in Bay Area metros, you may be overconcentrating, losing diversity, and getting groupthink in the algorithmic economy. It locks in a winner-take-most dimension to this sector, and that's where we hope that federal policy will begin to invest in new and different AI clusters in new and different places to provide a balance or counter," Mark Muro, policy director at the Brookings Institution and the study's coauthor, told WIRED.

This week you'll be able to listen to the talks of Jonathan Hurst (Professor of Robotics at Oregon State University, and Chief Technology Officer at Agility Robotics) and Andrea Thomaz (Associate Professor of Robotics at the University of Texas at Austin, and CEO of Diligent Robotics) as part of this series that brings you the plenary and keynote talks from the IEEE/RSJ IROS2020 (International Conference on Intelligent Robots and Systems). Jonathan's talk is in the topic of humanoids, while Andrea's is about human-robot interaction. Bio: Jonathan W. Hurst is Chief Technology Officer and co-founder of Agility Robotics, and Professor and co-founder of the Oregon State University Robotics Institute. He holds a B.S. in mechanical engineering and an M.S. and Ph.D. in robotics, all from Carnegie Mellon University. His university research focuses on understanding the fundamental science and engineering best practices for robotic legged locomotion and physical interaction.

BANGKOK/LONDON – A week after the Myanmar military seized power, a Twitter account that had lain dormant for nearly a decade flickered back into life. The Twitter user mocked anti-coup protesters, hundreds of whom have been killed in a crackdown by security forces since the Feb. 1 coup. After a police truck fired high-pressure water cannons on demonstrators in the capital city of Naypyidaw on Feb. 8, he made a trolling reference to the nation's traditional April new year celebration: "Water festival come earlier for them lol." A few weeks later, the user wrote "#fuckthereds," making a dismissive reference to the political party of Aung San Suu Kyi, the Nobel Prize-winning civilian leader who had been overthrown and arrested in the coup. A review of an archived version of the account, which has since been shut down, revealed the username was a pseudonym belonging to Ivan Htet, the 33-year-old son of a leading figure in the coup: the chief of the air force, Maung Maung Kyaw. But Ivan Htet hasn't just been an enthusiastic supporter on social media of the Tatmadaw, the name for the Myanmar military, which has dominated political life since independence in 1948 for Myanmar, then called Burma. He is also trying to cash in, helping equip the military, along with his wife Lin Lett Thiri, who co-founded a private firm to supply Myanmar's armed forces, Reuters has found.