With the rapid growth of machine learning, deep neural networks (DNNs) are now being used in numerous domains. Unfortunately, DNNs are "black-boxes", and cannot be interpreted by humans, which is a substantial concern in safety-critical systems. To mitigate this issue, researchers have begun working on explainable AI (XAI) methods, which can identify a subset of input features that are the cause of a DNN's decision for a given input. Most existing techniques are heuristic, and cannot guarantee the correctness of the explanation provided. In contrast, recent and exciting attempts have shown that formal methods can be used to generate provably correct explanations. Although these methods are sound, the computational complexity of the underlying verification problem limits their scalability; and the explanations they produce might sometimes be overly complex. Here, we propose a novel approach to tackle these limitations. We (1) suggest an efficient, verification-based method for finding minimal explanations, which constitute a provable approximation of the global, minimum explanation; (2) show how DNN verification can assist in calculating lower and upper bounds on the optimal explanation; (3) propose heuristics that significantly improve the scalability of the verification process; and (4) suggest the use of bundles, which allows us to arrive at more succinct and interpretable explanations. Our evaluation shows that our approach significantly outperforms state-of-the-art techniques, and produces explanations that are more useful to humans. We thus regard this work as a step toward leveraging verification technology in producing DNNs that are more reliable and comprehensible.

Achieving distributed reinforcement learning (RL) for large-scale cooperative multi-agent systems (MASs) is challenging because: (i) each agent has access to only limited information; (ii) issues on convergence or computational complexity emerge due to the curse of dimensionality. In this paper, we propose a general computationally efficient distributed framework for cooperative multi-agent reinforcement learning (MARL) by utilizing the structures of graphs involved in this problem. We introduce three coupling graphs describing three types of inter-agent couplings in MARL, namely, the state graph, the observation graph and the reward graph. By further considering a communication graph, we propose two distributed RL approaches based on local value-functions derived from the coupling graphs. The first approach is able to reduce sample complexity significantly under specific conditions on the aforementioned four graphs. The second approach provides an approximate solution and can be efficient even for problems with dense coupling graphs. Here there is a trade-off between minimizing the approximation error and reducing the computational complexity. Simulations show that our RL algorithms have a significantly improved scalability to large-scale MASs compared with centralized and consensus-based distributed RL algorithms.

Southwest Airlines Chief Operating Officer Andrew Watterson will apologize on Thursday before a U.S. Senate committee over the holiday meltdown that led to the cancellation of 16,700 flights and pledge changes to ensure that there will be no repeats. "Let me be clear: we messed up. In hindsight, we did not have enough winter operational resilience," Watterson's written testimony for a U.S. Senate Commerce Committee hearing seen by Reuters says. In other written testimony seen by Reuters, Southwest Airlines Pilots Association (SWAPA) President Casey Murray will tell the committee that the low-cost carrier's "overconfidence" in planning and a "systemic failure to provide modern tools" were responsible for the December meltdown that the union said stranded 2 million passengers and is estimated to have cost it more than $1 billion. Murray will tell the committee that pilots "have been sounding the alarm about (Southwest's) inadequate crew scheduling technology and outdated operational processes for years. Unfortunately, those warnings were summarily ignored."

Three months ago, ChatGPT debuted--the first artificial-intelligence bot to produce original content virtually indistinguishable from that of a human brain. Now, the creators of that software are beta testing a new tool that can (or so they say) determine whether a text was written by a person or a machine. The applications could be many, from identifying disinformation campaigns to detecting when a job candidate is has used AI for a cover letter. But experts worry the software will only create more challenges for leaders already caught in an AI rabbit hole. "The mushing of original thinking and discernment and artificial intelligence is dangerous for employees, managers, and leaders," says Andrés Tapia, a senior client partner and global diversity, equity, and inclusion strategist at Korn Ferry.

Fox News Flash top headlines are here. Check out what's clicking on Foxnews.com. Firefighters worked for a second day Wednesday to fully extinguish a blaze at a U.S. company's drone plant in Latvia. Local police said nothing had been found so far to indicate sabotage. Latvia's State Fire and Rescue Service was alerted Tuesday afternoon that a fire had broken out at Edge Autonomy's drone production plant in Marupe, a town that borders the capital, Riga. The Baltic News Service said that although the blaze was largely contained by 7 p.m. on Tuesday, firefighters continued work to fully extinguish the fire early Wednesday.\

AI-generated text, from tools like ChatGPT, is starting to impact daily life. Teachers are testing it out as part of classroom lessons. Marketers are champing at the bit to replace their interns. Memers are going buck wild. It would be a lie to say I'm not a little anxious about the robots coming for my writing gig.

AI chatbot ChatGPT has taken the world by storm and reached more than 100 million users just three months after launching in November. The AI bot, created by San Francisco-based company OpenAI, has been trained on a massive amount of text so it can generate human-like text in response to questions. But the popular technology has now been accused of being'woke' after a string of responses displaying a heavy left-wing bias, including refusing to praise Donald Trump or argue in favour of fossil fuels. ChatGPT said praising the former US President was'not appropriate' despite complimenting President Joe Biden's'knowledge, experience and vision'. It also wouldn't tell a joke about women as doing so would be'offensive or inappropriate', but happily told a joke about men.

Don't miss out on ET Prime stories! Get your daily dose of business updates on WhatsApp. Adani Group stocks rallied in early trade on Tuesday after the promoters said they had repaid ₹9,250 crore ($1.1 billion) of debt backed by shares in three companies. The home and IT ministries, along with the Reserve Bank of India (RBI), will hold discussions over the next few days to decide on the next steps after the government blocked 94 online lending apps earlier this week, multiple people aware of the developments said. India's food safety regulator is likely to mandate online marketplaces to prominently display key health and nutritional alerts alongside the retail price for packaged items sold on their platforms, multiple people aware of the matter told ET.

The "news broadcasters" appear stunningly real, but they are AI-generated deepfakes in first-of-their-kind propaganda videos that a research report published Tuesday attributed to Chinese state-aligned actors. The fake anchors -- for a fictious news outlet called Wolf News -- were created by artificial intelligence software and appeared in footage on social media that seemed to promote the interests of the Chinese Communist Party, US-based research firm Graphika said in its report. "This is the first time we've seen a state-aligned operation use AI-generated video footage of a fictitious person to create deceptive political content," Jack Stubbs, vice president of intelligence at Graphika, told AFP. In one video analyzed by Graphika, a fictious male anchor who calls himself Alex critiques US inaction over gun violence plaguing the country. In the second, a female anchor stresses the importance of "great power cooperation" between China and the United States.

Why should moral philosophers, moral psychologists, and machine ethicists care about computational complexity? Debates on whether artificial intelligence (AI) can or should be used to solve problems in ethical domains have mainly been driven by what AI can or cannot do in terms of human capacities. In this paper, we tackle the problem from the other end by exploring what kind of moral machines are possible based on what computational systems can or cannot do. To do so, we analyze normative ethics through the lens of computational complexity. First, we introduce computational complexity for the uninitiated reader and discuss how the complexity of ethical problems can be framed within Marr's three levels of analysis. We then study a range of ethical problems based on consequentialism, deontology, and virtue ethics, with the aim of elucidating the complexity associated with the problems themselves (e.g., due to combinatorics, uncertainty, strategic dynamics), the computational methods employed (e.g., probability, logic, learning), and the available resources (e.g., time, knowledge, learning). The results indicate that most problems the normative frameworks pose lead to tractability issues in every category analyzed. Our investigation also provides several insights about the computational nature of normative ethics, including the differences between rule- and outcome-based moral strategies, and the implementation-variance with regard to moral resources. We then discuss the consequences complexity results have for the prospect of moral machines in virtue of the trade-off between optimality and efficiency. Finally, we elucidate how computational complexity can be used to inform both philosophical and cognitive-psychological research on human morality by advancing the Moral Tractability Thesis (MTT).