Please use the sharing tools found via the share button at the top or side of articles. Subscribers may share up to 10 or 20 articles per month using the gift article service. More information can be found here. The writer is professor of computer science and Smith-Zadeh professor in engineering, University of California, Berkeley The subject of autonomous killer robots exercises many technologists, politicians and human rights activists. Indeed, the Financial Times's advice page for would-be opinion writers complains that, in their pitches, "lots of people spin doomsday scenarios about robots".

In the past few years, digital initiatives like making the internet more accessible, boosting IoT, cybersecurity, machine learning, and artificial intelligence in India have been the goals of the government. You must have heard that AI is the future and it particularly saves the world as it is a technology that will dramatically alter human life in very real ways. AI helps people to rethink how they integrate information, analyze data, and use the resulting insights to improve decision-making. With AI significantly changing the tech scenario across the globe, maybe conveying it successfully, and ensuring that its advantages stream down to the most minimal level, turns out to be particularly essential. According to reports, the number of smartphone users in India will reach over 760 million in the year 2021.

The third series of Global Information Dominance Experiments (GIDE) was held at Peterson Air Force Base, Colorado, on July 13, 2021. Before we talk about the US military using big data and artificial intelligence to try to predict future events, we might as well address the elephant -- or rather the notoriously small-statured but undeniably charismatic actor -- in the room. Yes, it sounds a lot like that old 2002 Tom Cruise sci-fi classic Minority Report; the one in which law enforcement uses genetically mutated human "precogs" with psychic abilities to bust criminals before they actually commit their crimes. "What we've seen is the ability to get way further -- what I call left -- left of being reactive to actually being proactive," Gen. Glen D. VanHerck, commander of North American Aerospace Defense Command, or NORAD, and US Northern Command told reporters at a briefing last week. VanHerck was discussing the latest results of the Global Information Dominance Experiment, also known as GIDE, a so-called cross-command event that involved representatives from all 11 combatant commands in the US Department of Defense.

The Department of Defense Joint Artificial Intelligence Center (JAIC) is continuing to operationalize Responsible Artificial Intelligence with the announcement of a pilot of a procurement review process that will ensure AI acquired by the JAIC is aligned with DoD's AI Ethics Principles. The DoD AI Ethics Principles, along with operational commitments, are outlined in the recently issued memorandum on implementing responsible AI. Responsible AI Institute will support the pilot. "The development of a tactical responsible AI procurement process is essential for creating the necessary safeguards aligned with our AI Ethics Principles and further implement against the responsible AI tenants set forth by the Deputy Secretary of Defense," says Alka Patel, Chief of Responsible AI at the JAIC. The Responsible AI Procurement pilot is part of a holistic approach that focuses not only on the technology but also on organizational operating structures and culture to advance Responsible AI within the DoD.

Jeremy Kepner, a Lincoln Laboratory Fellow in the Cyber Security and Information Sciences Division and a research affiliate of the MIT Department of Mathematics, was named to the 2021 class of fellows of the Society for Industrial and Applied Mathematics (SIAM). The fellow designation honors SIAM members who have made outstanding contributions to the 17 mathematics-related research areas that SIAM promotes through its publications, conferences, and community of scientists. Kepner was recognized for "contributions to interactive parallel computing, matrix-based graph algorithms, green supercomputing, and big data." Since joining Lincoln Laboratory in 1998, Kepner has worked to expand the capabilities of computing at the laboratory and throughout the computing community. He has published broadly, served on technical committees of national conferences, and contributed to regional efforts to provide access to supercomputing.

The BCS has published a report advocating that the UK should take a global lead on the application of ethics to artificial intelligence (AI). It said that the UK "can lead the world in creating AI that cares about humanity – provided more people from non-tech backgrounds choose the field". It added that a more diverse range of people need to be in AI-related jobs if public trust in the technology is to grow, and that government and the IT industry need to join forces to make sure that happens. The policy discussion document Priorities for the national AI strategy is intended to complement the UK government's National AI Strategy, which is due to be released later this year, a BCS spokesperson said. The Department for Digital, Culture, Media and Sport (DCMS) and the government's Office for AI are aware of the contents of the BCS document.

Embodied AI refers to the study and development of artificial agents that can perceive, reason, and interact with the environment with the capabilities and limitations of a physical body. Recently, significant progress has been made in developing solutions to embodied AI problems such as (visual) navigation [1-5], interactive Q&A [6-10], instruction following [11-15], and manipulation [16-22]. To calibrate the progress, several lines of pioneering efforts have been made towards benchmarking embodied AI in simulated environments, including Rearrangement [23, 24], TDW Transport Challenge [25], VirtualHome [26], ALFRED [11], Interactive Gibson Benchmark [27], MetaWorld [28], and RLBench [29], among others [30-32]). These efforts are inspiring, but their activities represent only a fraction of challenges that humans face in their daily lives. To develop artificial agents that can eventually perform and assist with everyday activities with human-level robustness and flexibility, we need a comprehensive benchmark with activities that are more realistic, diverse, and complex. But this is easier said than done. There are three major challenges that have prevented existing benchmarks to accommodate more realistic, diverse, and complex activities: - Definition: Identifying and defining meaningful activities for benchmarking; - Realization: Developing simulated environments that realistically support such activities; - Evaluation: Defining success and objective metrics for evaluating performance.

Imitating human demonstrations is a promising approach to endow robots with various manipulation capabilities. While recent advances have been made in imitation learning and batch (offline) reinforcement learning, a lack of open-source human datasets and reproducible learning methods make assessing the state of the field difficult. In this paper, we conduct an extensive study of six offline learning algorithms for robot manipulation on five simulated and three real-world multi-stage manipulation tasks of varying complexity, and with datasets of varying quality. Our study analyzes the most critical challenges when learning from offline human data for manipulation. Based on the study, we derive a series of lessons including the sensitivity to different algorithmic design choices, the dependence on the quality of the demonstrations, and the variability based on the stopping criteria due to the different objectives in training and evaluation. We also highlight opportunities for learning from human datasets, such as the ability to learn proficient policies on challenging, multi-stage tasks beyond the scope of current reinforcement learning methods, and the ability to easily scale to natural, real-world manipulation scenarios where only raw sensory signals are available. We have open-sourced our datasets and all algorithm implementations to facilitate future research and fair comparisons in learning from human demonstration data. Codebase, datasets, trained models, and more available at https://arise-initiative.github.io/robomimic-web/

As machine learning models grow more complex and their applications become more high-stakes, tools for explaining model predictions have become increasingly important. This has spurred a flurry of research in model explainability and has given rise to feature attribution methods such as LIME and SHAP. Despite their widespread use, evaluating and comparing different feature attribution methods remains challenging: evaluations ideally require human studies, and empirical evaluation metrics are often data-intensive or computationally prohibitive on real-world datasets. In this work, we address this issue by releasing XAI-Bench: a suite of synthetic datasets along with a library for benchmarking feature attribution algorithms. Unlike real-world datasets, synthetic datasets allow the efficient computation of conditional expected values that are needed to evaluate ground-truth Shapley values and other metrics. The synthetic datasets we release offer a wide variety of parameters that can be configured to simulate real-world data. We demonstrate the power of our library by benchmarking popular explainability techniques across several evaluation metrics and across a variety of settings. The versatility and efficiency of our library will help researchers bring their explainability methods from development to deployment. Our code is available at https://github.com/abacusai/xai-bench.

Israel's defense minister warned Thursday that his country is prepared to strike Iran, issuing the threat against the Islamic Republic after a fatal drone strike on a oil tanker at sea that his nation blamed on Tehran. The comments by Benny Gantz come as Israel lobbies countries for action at the United Nations over last week's attack on the oil tanker Mercer Street that killed two people. The tanker, struck off Oman in the Arabian Sea, is managed by a firm owned by an Israeli billionaire. The U.S. and the United Kingdom also blamed Iran for the attack, but no country has offered evidence or intelligence to support the claim. Iran, which along with its regional militia allies has launched similar drone attacks, has denied being involved.