Convolutional Neural Networks (CNNs) are frequently and successfully used in medical prediction tasks. They are often used in combination with transfer learning, leading to improved performance when training data for the task are scarce. The resulting models are highly complex and typically do not provide any insight into their predictive mechanisms, motivating the field of 'explainable' artificial intelligence (XAI). However, previous studies have rarely quantitatively evaluated the 'explanation performance' of XAI methods against ground-truth data, and transfer learning and its influence on objective measures of explanation performance has not been investigated. Here, we propose a benchmark dataset that allows for quantifying explanation performance in a realistic magnetic resonance imaging (MRI) classification task. We employ this benchmark to understand the influence of transfer learning on the quality of explanations. Experimental results show that popular XAI methods applied to the same underlying model differ vastly in performance, even when considering only correctly classified examples. We further observe that explanation performance strongly depends on the task used for pre-training and the number of CNN layers pre-trained. These results hold after correcting for a substantial correlation between explanation and classification performance.

Differentially Private Stochastic Gradient Descent (DP-SGD) limits the amount of private information deep learning models can memorize during training. This is achieved by clipping and adding noise to the model's gradients, and thus networks with more parameters require proportionally stronger perturbation. As a result, large models have difficulties learning useful information, rendering training with DP-SGD exceedingly difficult on more challenging training tasks. Recent research has focused on combating this challenge through training adaptations such as heavy data augmentation and large batch sizes. However, these techniques further increase the computational overhead of DP-SGD and reduce its practical applicability. In this work, we propose using the principle of sparse model design to solve precisely such complex tasks with fewer parameters, higher accuracy, and in less time, thus serving as a promising direction for DP-SGD. We achieve such sparsity by design by introducing equivariant convolutional networks for model training with Differential Privacy. Using equivariant networks, we show that small and efficient architecture design can outperform current state-of-the-art models with substantially lower computational requirements. On CIFAR-10, we achieve an increase of up to $9\%$ in accuracy while reducing the computation time by more than $85\%$. Our results are a step towards efficient model architectures that make optimal use of their parameters and bridge the privacy-utility gap between private and non-private deep learning for computer vision.

The president speaks after meeting with AI experts in effort to manage its risks. President Biden is expected to discuss artificial intelligence Tuesday afternoon in San Francisco in his first public speech since son Hunter Biden signed a plea deal on federal tax charges. Hunter Biden will plead guilty to two misdemeanor counts of willful failure to pay federal income tax, Fox News learned Tuesday. "Despite owing in excess of $100,000 in federal income taxes each year, he did not pay the income tax due for either year," the U.S. Attorney for the District of Delaware David C. Weiss' office said. He will also enter into a pretrial diversion agreement regarding a separate felony charge of possession of a firearm by a person who is an unlawful user of or addicted to a controlled substance.

New York City grocers are expressing outrage over a push by city council members to ban facial recognition technology stores rely on to deter shoplifting due to concerns of racial discrimination. Ferreira Foodtown CEO Jason Ferreira joined "Fox & Friends" Tuesday to call out the suggestion as thefts continue to rock businesses in the Big Apple. Ferreira, who has been in business for over 45 years, said the shoplifting has never been worse. "It's not only people that are doing it professionally. We have people that are doing it just because they can get away with it. And the gamut runs from children to people that are older."

The wheels of government have finally begun to turn on the issue of generative AI regulation. US Representatives Ted Lieu (D-CA) and Ken Buck (R-CO) introduced legislation on Monday that would establish a 20-person commission to study ways to "mitigate the risks and possible harms" of AI while "protecting" America's position as a global technology power. The bill would require the Executive branch to appoint experts from throughout government, academia and industry to conduct the study over the course of two years, producing three reports during that period. The president would appoint eight members of the committee, while Congress, in an effort "to ensure bipartisanship," would split the remaining 12 positions evenly between the two parties (thereby ensuring the entire process devolves into a partisan circus). "[Generative AI] can be disruptive to society, from the arts to medicine to architecture to so many different fields, and it could also potentially harm us and that's why I think we need to take a somewhat different approach," Lieu told the Washington Post.

Former President Donald Trump defends himself against allegations he mishandled classified documents on'Special Report.' Former President Donald Trump's trial on 37 federal felony counts is poised to begin on August 14, a judge announced Tuesday. Federal Judge Aileen Cannon announced the preliminary court date Tuesday, but the final date for Trump's trial is likely to change as the former president's legal team is expected to request a delay. Trump has vowed to continue his 2024 presidential campaign despite his legal jeopardy. Trump is accused of 37 counts, including willful retention of national defense information, conspiracy to obstruct justice and making false statements.

Haywood Talcove, CEO of LexisNexis Risk Solutions' government division, told Fox News Digital that he believes there will be more than $1 trillion in artificial intelligence-assisted fraud if U.S. doesn't act quickly. Artificial intelligence smashed the floodgates to unprecedented fraud that could cost taxpayers hundreds of billions, if not $1 trillion, over the next 12 months, an expert told Fox News Digital. Haywood Talcove, CEO of LexisNexis Risk Solutions' government division, which evaluates and predicts risk, said he's already seeing criminals on the dark web using people's faces to steal from government and state agencies. Benefits to America's most vulnerable communities, such as Social Security, Medicare and Medicaid, and unemployment, are ending up in the pockets of criminals and criminal enterprises that are operating all over the world. "Being one of the wealthiest countries in the world makes us a huge target," Talcove said.

The CEO of OpenAI, Sam Altman, has spent the last month touring world capitals where, at talks to sold-out crowds and in meetings with heads of governments, he has repeatedly spoken of the need for global AI regulation. But behind the scenes, OpenAI has lobbied for significant elements of the most comprehensive AI legislation in the world--the E.U.'s AI Act--to be watered down in ways that would reduce the regulatory burden on the company, according to documents about OpenAI's engagement with E.U. officials obtained by TIME from the European Commission via freedom of information requests. In several cases, OpenAI proposed amendments that were later made to the final text of the E.U. law--which was approved by the European Parliament on June 14, and will now proceed to a final round of negotiations before being finalized as soon as January. In 2022, OpenAI repeatedly argued to European officials that the forthcoming AI Act should not consider its general purpose AI systems--including GPT-3, the precursor to ChatGPT, and the image generator Dall-E 2--to be "high risk," a designation that would subject them to stringent legal requirements including transparency, traceability, and human oversight. That argument brought OpenAI in line with Microsoft, which has invested $13 billion into the AI lab, and Google, both of which have previously lobbied E.U. officials in favor of loosening the Act's regulatory burden on large AI providers.

Fueled by the soaring popularity of large language and foundation models, the accelerated growth of artificial intelligence (AI) models' enormous environmental footprint has come under increased scrutiny. While many approaches have been proposed to make AI more energy-efficient and environmentally friendly, environmental inequity -- the fact that AI's environmental footprint can be disproportionately higher in certain regions than in others -- has emerged, raising social-ecological justice concerns. This paper takes a first step toward addressing AI's environmental inequity by balancing its regional negative environmental impact. Concretely, we focus on the carbon and water footprints of AI model inference and propose equity-aware geographical load balancing (GLB) to explicitly address AI's environmental impacts on the most disadvantaged regions. We run trace-based simulations by considering a set of 10 geographically-distributed data centers that serve inference requests for a large language AI model. The results demonstrate that existing GLB approaches may amplify environmental inequity while our proposed equity-aware GLB can significantly reduce the regional disparity in terms of carbon and water footprints.

Artificial Intelligence (AI) has made impressive progress in recent years and represents a key technology that has a crucial impact on the economy and society. However, it is clear that AI and business models based on it can only reach their full potential if AI applications are developed according to high quality standards and are effectively protected against new AI risks. For instance, AI bears the risk of unfair treatment of individuals when processing personal data e.g., to support credit lending or staff recruitment decisions. The emergence of these new risks is closely linked to the fact that the behavior of AI applications, particularly those based on Machine Learning (ML), is essentially learned from large volumes of data and is not predetermined by fixed programmed rules. Thus, the issue of the trustworthiness of AI applications is crucial and is the subject of numerous major publications by stakeholders in politics, business and society. In addition, there is mutual agreement that the requirements for trustworthy AI, which are often described in an abstract way, must now be made clear and tangible. One challenge to overcome here relates to the fact that the specific quality criteria for an AI application depend heavily on the application context and possible measures to fulfill them in turn depend heavily on the AI technology used. Lastly, practical assessment procedures are needed to evaluate whether specific AI applications have been developed according to adequate quality standards. This AI assessment catalog addresses exactly this point and is intended for two target groups: Firstly, it provides developers with a guideline for systematically making their AI applications trustworthy. Secondly, it guides assessors and auditors on how to examine AI applications for trustworthiness in a structured way.