Bipartisan legislation introduced on Thursday would direct the Defense Department to make libraries of its vast datasets publicly available and readily accessible via a strategic pilot program aimed at accelerating software and artificial intelligence capabilities development.

Local governments in Japan are turning to artificial intelligence to improve communication with people who are deaf or hard of hearing at their counters for the public. A system jointly developed by the University of Electro-Communications in Tokyo and SoftBank Corp. converts sign-language gestures into written text. While the system currently requires equipment at counters, municipalities hope it will eventually be usable with just a smartphone. At the Narashino city office in Chiba Prefecture, a woman with a hearing disability asked directions to the restroom using sign language while standing in front of a camera. A text translation appeared on a staff member's computer display after about three seconds. The spoken response then appeared as text on the screen in front of the woman, making for a smooth interaction.

Monitoring of workers and setting performance targets through algorithms is damaging employees' mental health and needs to be controlled by new legislation, according to a group of MPs and peers. An "accountability for algorithms act'" would ensure that companies evaluate the effect of performance-driven regimes such as queue monitoring in supermarkets or deliveries-per-hour guidelines for delivery drivers, said the all-party parliamentary group (APPG) on the future of work. "Pervasive monitoring and target-setting technologies, in particular, are associated with pronounced negative impacts on mental and physical wellbeing as workers experience the extreme pressure of constant, real-time micro-management and automated assessment," said the APPG members in their report, the New Frontier: Artificial Intelligence at Work. The report recommends bringing in a new algorithms act, which it says would establish "a clear direction to ensure AI puts people first". It warns that "use of algorithmic surveillance, management and monitoring technologies that undertake new advisory functions, as well as traditional ones, has significantly increased during the pandemic".

The Department of Homeland Security (DHS) is collecting feedback and opinions regarding the use of artificial intelligence (AI) and facial recognition between now and December 6. DHS has used and piloted AI-enabled technologies in several functions like customs and border protection, transportation security, and investigations. Earlier this year, DHS launched new shoe-scanning imaging technology, to be deployed at TSA security checkpoints to improve the efficiency of airport screening and potentially eliminate the need to remove shoes and outerwear when passing through checkpoints. However, AI and facial recognition bring public controversies, such as bias, security, and privacy concerns. "Understanding how the public perceives these technologies and then designing and deploying them in a manner responsive to the public's concerns is critical in gaining public support for DHS's use of these technologies," an information collection request posted to the Federal Register stated.

The European Parliament should amend the regulation to better protect people's rights to social security and an adequate standard of living. The 28-page report in the form of a question-and-answer document, "How the EU's Flawed Artificial Intelligence Regulation Endangers the Social Safety Net," examines how governments are turning to algorithms to allocate social security support and prevent benefits fraud. Drawing on case studies in Ireland, France, the Netherlands, Austria, Poland, and the United Kingdom, Human Rights Watch found that this trend toward automation can discriminate against people who need social security support, compromise their privacy, and make it harder for them to qualify for government assistance. But the regulation will do little to prevent or rectify these harms. "The EU's proposal does not do enough to protect people from algorithms that unfairly strip them of the benefits they need to support themselves or find a job," said Amos Toh, senior researcher on artificial intelligence and human rights at Human Rights Watch.

Different agents need to make a prediction. They observe identical data, but have different models: they predict using different explanatory variables. We study which agent believes they have the best predictive ability -- as measured by the smallest subjective posterior mean squared prediction error -- and show how it depends on the sample size. With small samples, we present results suggesting it is an agent using a low-dimensional model. With large samples, it is generally an agent with a high-dimensional model, possibly including irrelevant variables, but never excluding relevant ones. We apply our results to characterize the winning model in an auction of productive assets, to argue that entrepreneurs and investors with simple models will be over-represented in new sectors, and to understand the proliferation of "factors" that explain the cross-sectional variation of expected stock returns in the asset-pricing literature.

Retrieving a signal from the Fourier transform of its third-order statistics or bispectrum arises in a wide range of signal processing problems. Conventional methods do not provide a unique inversion of bispectrum. In this paper, we present a an approach that uniquely recovers signals with finite spectral support (band-limited signals) from at least $3B$ measurements of its bispectrum function (BF), where $B$ is the signal's bandwidth. Our approach also extends to time-limited signals. We propose a two-step trust region algorithm that minimizes a non-convex objective function. First, we approximate the signal by a spectral algorithm. Then, we refine the attained initialization based upon a sequence of gradient iterations. Numerical experiments suggest that our proposed algorithm is able to estimate band/time-limited signals from its BF for both complete and undersampled observations.

The past decade has seen significant progress in artificial intelligence (AI), which has resulted in algorithms being adopted for resolving a variety of problems. However, this success has been met by increasing model complexity and employing black-box AI models that lack transparency. In response to this need, Explainable AI (XAI) has been proposed to make AI more transparent and thus advance the adoption of AI in critical domains. Although there are several reviews of XAI topics in the literature that identified challenges and potential research directions in XAI, these challenges and research directions are scattered. This study, hence, presents a systematic meta-survey for challenges and future research directions in XAI organized in two themes: (1) general challenges and research directions in XAI and (2) challenges and research directions in XAI based on machine learning life cycle's phases: design, development, and deployment. We believe that our meta-survey contributes to XAI literature by providing a guide for future exploration in the XAI area.

When technology enters applications and processes with a long tradition of controversial societal debate, multi-faceted new ethical and legal questions arise. This paper focusses on the process of whistleblowing, an activity with large impacts on democracy and business. Computer science can, for the first time in history, provide for truly anonymous communication. We investigate this in relation to the values and rights of accountability, fairness and data protection, focusing on opportunities and limitations of the anonymity that can be provided computationally; possible consequences of outsourcing whistleblowing support; and challenges for the interpretation and use of some relevant laws. We conclude that to address these questions, whistleblowing and anonymous whistleblowing must rest on three pillars, forming a 'triangle of whistleblowing protection and incentivisation' that combines anonymity in a formal and technical sense; whistleblower protection through laws; and organisational and political error culture.

Recent improvements in the performance of state-of-the-art (SOTA) methods for Graph Representational Learning (GRL) have come at the cost of significant computational resource requirements for training, e.g., for calculating gradients via backprop over many data epochs. Meanwhile, Singular Value Decomposition (SVD) can find closed-form solutions to convex problems, using merely a handful of epochs. In this paper, we make GRL more computationally tractable for those with modest hardware. We design a framework that computes SVD of \textit{implicitly} defined matrices, and apply this framework to several GRL tasks. For each task, we derive linear approximation of a SOTA model, where we design (expensive-to-store) matrix $\mathbf{M}$ and train the model, in closed-form, via SVD of $\mathbf{M}$, without calculating entries of $\mathbf{M}$. By converging to a unique point in one step, and without calculating gradients, our models show competitive empirical test performance over various graphs such as article citation and biological interaction networks. More importantly, SVD can initialize a deeper model, that is architected to be non-linear almost everywhere, though behaves linearly when its parameters reside on a hyperplane, onto which SVD initializes. The deeper model can then be fine-tuned within only a few epochs. Overall, our procedure trains hundreds of times faster than state-of-the-art methods, while competing on empirical test performance. We open-source our implementation at: https://github.com/samihaija/isvd