Despite substantial progress in applying neural networks (NN) to a wide variety of areas, they still largely suffer from a lack of transparency and interpretability. While recent developments in explainable artificial intelligence attempt to bridge this gap (e.g., by visualizing the correlation between input pixels and final outputs), these approaches are limited to explaining low-level relationships, and crucially, do not provide insights on error correction. In this work, we propose a framework (VRX) to interpret classification NNs with intuitive structural visual concepts. Given a trained classification model, the proposed VRX extracts relevant class-specific visual concepts and organizes them using structural concept graphs (SCG) based on pairwise concept relationships. By means of knowledge distillation, we show VRX can take a step towards mimicking the reasoning process of NNs and provide logical, concept-level explanations for final model decisions. With extensive experiments, we empirically show VRX can meaningfully answer "why" and "why not" questions about the prediction, providing easy-to-understand insights about the reasoning process. We also show that these insights can potentially provide guidance on improving NN's performance.

In one 2020 complaint to the Consumer Financial Protection Bureau, a consumer echoed the words of George Floyd to describe an experience with a financial company, saying "you all will not let me breathe." The consumer wanted to know why the firm would "not take their knee off ... my neck?" Another criticized a company for its approach to sexual identity issues. "The employees refused to be sensitive to my pronouns' and name change," the consumer said. "As a result, my account was closed after years of torture from this credit card company."

NASA's Perseverance rover captured this image of the Ingenuity helicopter during its fourth flight The Ingenuity helicopter has flown on Mars once again. During its fourth flight on 30 April, the drone flew further and faster than ever before – and it is far from finished, as NASA also announced a month-long extension to its mission. Ingenuity achieved all of the main goals of its mission in its first three flights, so the aim of the fourth flight was to push beyond that slightly. The fourth flight lasted nearly two minutes, during which the helicopter flew 266 metres and reached a top speed of 13 kilometres per hour. The mission was planned to include only five flights before the Perseverance rover drove away and began its own primary missions of searching for signs of ancient life and collecting samples to be returned to Earth.

This story was published in partnership with Type Investigations with support from the Puffin Foundation. In California, non-English speakers handed COVID-19 vaccination cards without information on what they mean. In Pennsylvania, people who speak Mandarin, Korean, and Japanese unable to make vaccine appointments due to a lack of interpreters at hospital call centers. These are just a few of the examples captured in a new complaint filed on Friday to the U.S. Department of Health and Human Services' Office for Civil Rights, Federal Emergency Management Agency's Office of Equal Rights, and Department of Homeland Security's Office for Civil Rights and Civil Liberties. The complaint, brought by the National Health Law Program, finds widespread problems across the country that inhibit access to COVID-19 resources for people with limited English proficiency (LEP).

A written proposal to ban several uses of artificial intelligence (AI) and to place new oversight on other "high-risk" AI applications--published by the European Commission this week--met fierce opposition from several digital rights advocates in Europe. Portrayed as a missed opportunity by privacy experts, the EU Commission's proposal bans four broad applications of AI, but it includes several loopholes that could lead to abuse, and it fails to include a mechanism to add other AI applications to the ban list. It deems certain types of AI applications as "high-risk"--meaning their developers will need to abide by certain restrictions--but some of those same applications were specifically called out by many digital rights groups earlier this year as "incompatible with a democratic society." It creates new government authorities, but the responsibilities of those authorities may overlap with separate authorities devoted to overall data protection. Most upsetting to digital rights experts, it appears, is that the 107-page document (not including the necessary annexes) offers only glancing restrictions on biometric surveillance, like facial recognition software.

The Tianhe-1 module that launched this week is the core of what is supposed to be a three-part space station. On the surface, it seems to pale in comparison to the 22-year-old ISS. The ISS is a football-field-size behemoth weighing about 420 metric tons, while the much smaller T-shaped Chinese Space Station (CSS) will be a mere 80 to 100 tons, closer to the size and mass of Russia's former Mir station. The Tianhe-1 module is just 22 tons and 16.6 meters long. And after 12 missions this year and next to put the whole thing together, the completed station will still be roughly half the length of the ISS.

While US lawmakers muddle through yet another congressional hearing on the dangers posed by algorithmic bias in social media, the European Commission (basically the executive branch of the EU) has unveiled a sweeping regulatory framework that, if adopted, could have global implications for the future of AI development. After extensive meetings with advocate groups and other stakeholders, the EC released both the first European Strategy on AI and Coordinated Plan on AI in 2018. Those were followed in 2019 by the Guidelines for Trustworthy AI, then again in 2020 by the Commission's White Paper on AI and Report on the safety and liability implications of Artificial Intelligence, the Internet of Things and robotics. Just as with its ambitious General Data Protection Regulation (GDPR) plan in 2018, the Commission is seeking to establish a basic level of public trust in the technology based on strident user and data privacy protections as well as those against its potential misuse. "Artificial intelligence should not be an end in itself, but a tool that has to serve people with the ultimate aim of increasing human well-being. Rules for artificial intelligence available in the Union market or otherwise affecting Union citizens should thus put people at the centre (be human-centric), so that they can trust that the technology is used in a way that is safe and compliant with the law, including the respect of fundamental rights," the Commission included in its draft regulations.

The world's biggest airplane – longer than an American football field and with a wingspan of three Boeing 747s – completed its second test flight from Mojave Air and Space Port in California. Built by Stratolaunch, the plane named Roc – with a wingspan of 117 metres and powered by 6 Boeing 747 engines – is being groomed to carry and launch supersonic vehicles high up in the sky. During Thursday's three hour test flight, Roc reached an altitude of 4,267m and a top speed of 320 km/h. "We're very pleased with how the Stratolaunch aircraft performed today, and we are equally excited about how much closer the aircraft is to launching its first hypersonic vehicle," said Zachary Krevor, Stratolaunch chief operating officer, said in a postflight news conference. US army develops new tool to detect deepfakes threatening national security Apple broke competition law, EU says after Spotify complaint'Brain-like device' mimics human learning in major computing breakthrough'Brain-like device' mimics human learning in major computing breakthrough With greater ambitions of "convenient, affordable, and routine" access to space, Roc's successful test flight and immediate vision of carrying supersonic vehicles is a step in the right direction for the company which was established in 2011 by Microsoft co-founder Paul Allen.

Thus far, most attempts at making policy for artificial intelligence have fallen into one of two camps: either outright bans on certain applications of machine learning--for instance, the facial recognitions bans passed in a few cities in the United States--or very broad, high-level principles that offer no concrete guidance or specific rules, like the "Ethical Principles for Artificial Intelligence" that the Department of Defense adopted in 2020 for developing and implementing A.I. in a responsible, equitable, traceable, reliable, and governable manner. There are obvious drawbacks to both of these approaches: The former seems neither sustainable nor scalable, given the pace with which machine learning is advancing and the extent to which both public and private entities seem eager to adopt it, while the latter often amounts to little more than window-dressing and vague reassurances that policymakers are at least thinking about the big questions posed by automated decision-making. That's why it was such a big deal when the European Commission released its proposed rules for artificial intelligence last week. While the rules are not yet final, the 108-page document "laying down harmonised rules on artificial intelligence" is certainly the closest any regulatory body has come to trying to develop more detailed and nuanced rules for A.I. than just banning certain uses or promise to implement these systems in generally ethical ways. European Commission executive vice president Margrethe Vestager hailed the new document as a set of "landmark rules" and others have since referred to them as "strict" or "ambitious," or an effort by the EU to become a "super-regulator."

Servers run inside the Facebook New Albany Data Center on Thursday, February 6, 2020 in New Albany, Ohio. Artificial intelligence (AI) is causing significant structural changes to global competition and economic growth. AI may generate trillions of dollars in new value over the next decade, but this value will not be easily captured or evenly distributed across nations. Much of it will depend on how governments invest in the underlying computational infrastructure that makes AI possible. Yet early signs point to a blind spot--a lack of understanding, measurement, and planning.