Due to their convenience and high accuracy, face recognition systems are widely employed in governmental and personal security applications to automatically recognise individuals. Despite recent advances, face recognition systems have shown to be particularly vulnerable to identity attacks (i.e., digital manipulations and attack presentations). Identity attacks pose a big security threat as they can be used to gain unauthorised access and spread misinformation. In this context, most algorithms for detecting identity attacks generalise poorly to attack types that are unknown at training time. To tackle this problem, we introduce a differential anomaly detection framework in which deep face embeddings are first extracted from pairs of images (i.e., reference and probe) and then combined for identity attack detection. The experimental evaluation conducted over several databases shows a high generalisation capability of the proposed method for detecting unknown attacks in both the digital and physical domains.

Electric Vehicles (EVs) are spreading fast as they promise to provide better performances and comfort, but above all, to help facing climate change. Despite their success, their cost is still a challenge. One of the most expensive components of EVs is lithium-ion batteries, which became the standard for energy storage in a wide range of applications. Precisely estimating the Remaining Useful Life (RUL) of battery packs can open to their reuse and thus help to reduce the cost of EVs and improve sustainability. A correct RUL estimation can be used to quantify the residual market value of the battery pack. The customer can then decide to sell the battery when it still has a value, i.e., before it exceeds its end of life of the target application and can still be reused in a second domain without compromising safety and reliability. In this paper, we propose to use a Deep Learning approach based on LSTMs and Autoencoders to estimate the RUL of li-ion batteries. Compared to what has been proposed so far in the literature, we employ measures to ensure the applicability of the method also in the real deployed application. Such measures include (1) avoid using non-measurable variables as input, (2) employ appropriate datasets with wide variability and different conditions, (3) do not use cycles to define the RUL.

The hydrodynamic performance of a sea-going ship varies over its lifespan due to factors like marine fouling and the condition of the anti-fouling paint system. In order to accurately estimate the power demand and fuel consumption for a planned voyage, it is important to assess the hydrodynamic performance of the ship. The current work uses machine-learning (ML) methods to estimate the hydrodynamic performance of a ship using the onboard recorded in-service data. Three ML methods, NL-PCR, NL-PLSR and probabilistic ANN, are calibrated using the data from two sister ships. The calibrated models are used to extract the varying trend in ship's hydrodynamic performance over time and predict the change in performance through several propeller and hull cleaning events. The predicted change in performance is compared with the corresponding values estimated using the fouling friction coefficient ($\Delta C_F$). The ML methods are found to be performing well while modelling the hydrodynamic state variables of the ships with probabilistic ANN model performing the best, but the results from NL-PCR and NL-PLSR are not far behind, indicating that it may be possible to use simple methods to solve such problems with the help of domain knowledge.

Probabilistic programming provides numerous advantages Ideally speaking, a good recommendations system should be over other techniques, including but not able to identify and remove malicious users before they can limited to providing a disentangled representation disrupt the ranking system by a significant margin. However, of how malicious users acted under a structured to eliminate the risk of false positives a resilient ranking model, as well as allowing for the quantification system can use as much data as possible. So we have to of damage caused by malicious users. We show adjust the tradeoff between false positives and the damage a experiments in malicious user identification using set of malicious users can cause to a ranking system.

What is learning? 20$^{st}$ century formalizations of learning theory -- which precipitated revolutions in artificial intelligence -- focus primarily on $\mathit{in-distribution}$ learning, that is, learning under the assumption that the training data are sampled from the same distribution as the evaluation distribution. This assumption renders these theories inadequate for characterizing 21$^{st}$ century real world data problems, which are typically characterized by evaluation distributions that differ from the training data distributions (referred to as out-of-distribution learning). We therefore make a small change to existing formal definitions of learnability by relaxing that assumption. We then introduce $\mathbf{learning\ efficiency}$ (LE) to quantify the amount a learner is able to leverage data for a given problem, regardless of whether it is an in- or out-of-distribution problem. We then define and prove the relationship between generalized notions of learnability, and show how this framework is sufficiently general to characterize transfer, multitask, meta, continual, and lifelong learning. We hope this unification helps bridge the gap between empirical practice and theoretical guidance in real world problems. Finally, because biological learning continues to outperform machine learning algorithms on certain OOD challenges, we discuss the limitations of this framework vis-\'a-vis its ability to formalize biological learning, suggesting multiple avenues for future research.

After experimenting with exoskeletons, performance-enhancing drugs, and AR goggles to develop super soldiers, the US military has taken a punt on a brain-modulating skullcap. The army has issued funding for a wearable device that analyzes how the brain disposes of waste during sleep. The system will be developed by researchers at Rice University, Houston Methodist Hospital, and Baylor College of Medicine. Ultimately, the team wants to create a headset that can treat sleep disorders in real-time. But first, they aim to track and adjust the flow of cerebrospinal fluid as it flushes waste out of the brain.

The European Commission announced first-of-its-kind legislation regulating the use of artificial intelligence in April. This unleashed criticism that the regulations could slow AI innovation, hamstringing Europe in its competition with the U.S. and China for leadership in AI. For example, Andrew McAfee wrote an article titled "EU proposals to regulate AI are only going to hinder innovation." Anticipating this criticism and mindful of the example of GDPR, where Europe's thought-leadership position didn't necessarily translate into data-related innovation, the EC has tried to address AI innovation directly by publishing a new Coordinated Plan on AI. Released in conjunction with the proposed regulations, the plan is full of initiatives intended to help the EU become a leader in AI technology.

The European Union has taken a further step towards banning the use of artificial intelligence to carry out mass surveillance, rule on court cases or predict whether individuals will commit crimes. New legislation that would introduce strict controls on AI to prevent racial, gender or age bias in particular "high risk" areas such as law enforcement is currently working its way through the European Parliament, but a report by the Committee on Civil Liberties, Justice and Home Affairs recently proposed even …

The European Parliament has called on lawmakers in the European Union to ban automated facial recognition in public spaces and to enforce strict safeguards for police use of artificial intelligence. MEPs voted in favor of the non-binding resolution by 377-248, with 62 abstentions. The MEPs said citizens should only be monitored when they're suspected of a crime. They cited concerns over algorithmic bias in AI and argued that both human supervision and legal protections are required to avoid discrimination. The politicians noted there's evidence suggesting AI-based identification systems misidentify minority ethnic groups, LGBTI people, seniors and women at higher rates.

Iranian dissidents are warning of the hard-line regime's use of drones to cause instability in the region, saying it is using the technology – materials for which are being imported from China – to make up for the weaknesses of its air force. The National Council of Resistance of Iran (NCRI), an umbrella group of Iranian resistance groups that oppose the regime, released evidence in a press conference it says shows the production and utilization of unmanned aerial vehicles (UACs) for terrorist operations and for assisting its proxies in the Middle East – including aerial photographs of the alleged sites and details that have emerged from inside the country. "Our revelation today is significant because it shows that the Qods Force of the IRGC has in recent years expanded its arsenal to step up terrorism and warmongering to destabilize the region by arming its proxies with UAVs," Alireza Jafarzadeh, deputy director of the Washington office of the National Council of Resistance of Iran, told Fox News. "This is in line with the regime's nuclear defiance and its repression at home." The group alleges that the regime, which has been rocked by a slew of economic sanctions imposed by the Trump administration as well as protests at home and challenges related to its handling of the COVID-19 pandemic, has used a web of industries to spend billions of dollars to produce components or smuggle them in from foreign countries.