Model brittleness is a key concern when deploying deep learning models in real-world medical settings. A model that has high performance at one institution may suffer a significant decline in performance when tested at other institutions. While pooling datasets from multiple institutions and re-training may provide a straightforward solution, it is often infeasible and may compromise patient privacy. An alternative approach is to fine-tune the model on subsequent institutions after training on the original institution. Notably, this approach degrades model performance at the original institution, a phenomenon known as catastrophic forgetting. In this paper, we develop an approach to address catastrophic forgetting based on elastic weight consolidation combined with modulation of batch normalization statistics under two scenarios: first, for expanding the domain from one imaging system's data to another imaging system's, and second, for expanding the domain from a large multi-institutional dataset to another single institution dataset. We show that our approach outperforms several other state-of-the-art approaches and provide theoretical justification for the efficacy of batch normalization modulation. The results of this study are generally applicable to the deployment of any clinical deep learning model which requires domain expansion.

From mid-January 2023 you can submit again in the category "Artificial Intelligence & Life Art" of the Prix Ars Electronica! The category „Artificial Intelligence & Life Art" is dedicated to artistic practice and thinking related with all areas of Artificial Intelligence and Life Sciences. We are also looking for artists who explore the intersections of these areas, such as robotics, androids, prosthetics or projects that address environmental issues, our bio-sphere and biodiversity. Art, science collaborations are of particular interest as are projects that critically reflect the cultural and social significance of AI and Life Sciences, their ethical and philosophical dimension as well as the role of policy makers, governments and the industry. The following materials are needed for the submission.

The world is facing enormous challenges, ranging from climate change to extreme poverty. The 2030 Agenda for Sustainable Development and its 17 Sustainable Development Goals (SDGs)a were adopted by United Nations Member States in 2015 as an operational framework to address these challenges. The SDGs include No Poverty, Quality Education, Gender Equality, Peace, Justice and Strong Institutions, among others, as well as a meta goal on Partnerships for the Goals. Despite limitations,7 the SDGs form a rare global consensus of all 193 UN member states on where we should collectively be heading. Goals are meaningless without a way to track their progress. Data on the SDGs and the associated indicatorsb are often outdated or unavailable, hindering progress during the Decade of Action leading up to 2030.c

Forty years ago, the word "hacker" was little known. Its march from obscurity to newspaper headlines owes a great deal to tech journalist Steven Levy, who in 1984 defied the advice of his publisher to call his first book Hackers: Heroes of the Computer Revolution.11 Hackers were a subculture of computer enthusiasts for whom programming was a vocation and playing around with computers constituted a lifestyle. Hackers was published only three years after Tracy Kidder's The Soul of a New Machine, explored in my last column (January 2021, p. 32–37), but a lot had changed during the interval. Kidder's assumed readers had never seen a minicomputer, still less designed one. By 1984, in contrast, the computer geek was a prominent part of popular culture. Unlike Kidder, Levy had to make people reconsider what they thought they already knew. Computers were suddenly everywhere, but they remained unfamiliar enough to inspire a host of popular books to ponder the personal and social transformations triggered by the microchip. The short-lived home computer boom had brought computer programming into the living rooms and basements of millions of middle-class Americans, sparking warnings about the perils of computer addiction. A satirical guide, published the same year, warned of "micromania."15 The year before, the film Wargames suggested computer-obsessed youth might accidentally trigger nuclear war.

Modern applications of survival analysis increasingly involve time-dependent covariates, which constitute a form of functional data. Learning from functional data generally involves repeated evaluations of time integrals which is numerically expensive. In this work we propose a lightweight data preprocessing step that transforms functional data into nonfunctional data. Boosting implementations for nonfunctional data can then be used, whereby the required numerical integration comes for free as part of the training phase. We use this to develop BoXHED 2.0, a quantum leap over the tree-boosted hazard package BoXHED 1.0. BoXHED 2.0 extends BoXHED 1.0 to Aalen's multiplicative intensity model, which covers censoring schemes far beyond right-censoring and also supports recurrent events data. It is also massively scalable because of preprocessing and also because it borrows from the core components of XGBoost. BoXHED 2.0 supports the use of GPUs and multicore CPUs, and is available from GitHub: www.github.com/BoXHED.

It is no secret that technology has been transforming the way we work. Every day, new tools and applications appear to automate processes, making them more agile and precise. In the area of finance, especially in the tax sector, the use of Artificial Intelligence has become increasingly fundamental to eliminate manual errors, increase productivity and make the business more strategic – mainly in the UK – where 1,958 hours are spent per year to meet all tax obligations, according to the World Bank. According to the survey conducted by Thomson Reuters in partnership with Live University, 56% of British companies intend to use Artificial Intelligence to optimize tax management. When the debate is about which technology is more functional for the sector, 61% of professionals point to Machine Learning as the innovation most capable of benefiting the segment; 31% bet on Data Science, and 10% prefer chatbots. The fact is that these combined technologies should revolutionize the finance area, as has already been happening with banks, in addition to other sectors such as e-commerce and general service, due to their high precision in collecting and data analysis, problem-solving, and responsiveness.

When a secretive start-up scraped the internet to build a facial-recognition tool, it tested a legal and ethical limit -- and blew the future of privacy in America wide open. In May 2019, an agent at the Department of Homeland Security received a trove of unsettling images. Found by Yahoo in a Syrian user's account, the photos seemed to document the sexual abuse of a young girl. One showed a man with his head reclined on a pillow, gazing directly at the camera. The man appeared to be white, with brown hair and a goatee, but it was hard to really make him out; the photo was grainy, the angle a bit oblique. The agent sent the man's face to child-crime investigators around the country in the hope that someone might recognize him. When an investigator in New York saw the request, she ran the face through an unusual new facial-recognition app she had just started using, called Clearview AI. The team behind it had scraped the public web -- social media, employment sites, YouTube, Venmo -- to create a database with three billion images of people, along with links to the webpages from which the photos had come. This dwarfed the databases of other such products for law enforcement, which drew only on official photography like mug shots, driver's licenses and passport pictures; with Clearview, it was effortless to go from a face to a Facebook account. The app turned up an odd hit: an Instagram photo of a heavily muscled Asian man and a female fitness model, posing on a red carpet at a bodybuilding expo in Las Vegas. The suspect was neither Asian nor a woman. But upon closer inspection, you could see a white man in the background, at the edge of the photo's frame, standing behind the counter of a booth for a workout-supplements company. On Instagram, his face would appear about half as big as your fingernail. The federal agent was astounded. The agent contacted the supplements company and obtained the booth worker's name: Andres Rafael Viola, who turned out to be an Argentine citizen living in Las Vegas.

Algorithmic decision-making (ADM) increasingly shapes people's daily lives. Given that such autonomous systems can cause severe harm to individuals and social groups, fairness concerns have arisen. A human-centric approach demanded by scholars and policymakers requires taking people's fairness perceptions into account when designing and implementing ADM. We provide a comprehensive, systematic literature review synthesizing the existing empirical insights on perceptions of algorithmic fairness from 39 empirical studies spanning multiple domains and scientific disciplines. Through thorough coding, we systemize the current empirical literature along four dimensions: (a) algorithmic predictors, (b) human predictors, (c) comparative effects (human decision-making vs. algorithmic decision-making), and (d) consequences of ADM. While we identify much heterogeneity around the theoretical concepts and empirical measurements of algorithmic fairness, the insights come almost exclusively from Western-democratic contexts. By advocating for more interdisciplinary research adopting a society-in-the-loop framework, we hope our work will contribute to fairer and more responsible ADM.

There has been a recent resurgence of interest in explainable artificial intelligence (XAI) that aims to reduce the opaqueness of AI-based decision-making systems, allowing humans to scrutinize and trust them. Prior work in this context has focused on the attribution of responsibility for an algorithm's decisions to its inputs wherein responsibility is typically approached as a purely associational concept. In this paper, we propose a principled causality-based approach for explaining black-box decision-making systems that addresses limitations of existing methods in XAI. At the core of our framework lies probabilistic contrastive counterfactuals, a concept that can be traced back to philosophical, cognitive, and social foundations of theories on how humans generate and select explanations. We show how such counterfactuals can quantify the direct and indirect influences of a variable on decisions made by an algorithm, and provide actionable recourse for individuals negatively affected by the algorithm's decision. Unlike prior work, our system, LEWIS: (1)can compute provably effective explanations and recourse at local, global and contextual levels (2)is designed to work with users with varying levels of background knowledge of the underlying causal model and (3)makes no assumptions about the internals of an algorithmic system except for the availability of its input-output data. We empirically evaluate LEWIS on three real-world datasets and show that it generates human-understandable explanations that improve upon state-of-the-art approaches in XAI, including the popular LIME and SHAP. Experiments on synthetic data further demonstrate the correctness of LEWIS's explanations and the scalability of its recourse algorithm.

Daniel E Ho and colleagues explore the legal implications of using artificial intelligence in the response to covid-19 and call for more robust evaluation frameworks Numerous proposals, prototypes, and models have emerged for using artificial intelligence (AI) and machine learning to predict individual risk related to covid-19. In the United States, for instance, the Department of Veterans Affairs uses individualised risk scores to allocate medical resources to people with covid-19,1 and prisons have sought to detect symptoms by processing inmates’ phone calls.2 Further tools, such as vulnerability predictions for individuals3 and voice based detection of infection,4 are on the horizon. But use of AI for such purposes has given rise to questions about legality. When a state or federal government seeks to use AI models to predict an individual’s risk of covid-19, the key legal questions will ultimately turn on how effective the models are and how much they burden legal interests. We focus on two of the most salient legal concerns under US law: privacy and discrimination. Challenges on privacy or discrimination grounds might appear in a variety of contexts, including challenges to regulatory decisions, tort actions, or lawsuits under health privacy laws. We argue that the basic need to balance benefits against burdens runs through all of these legal regimes. Governments implementing risk scoring tools must show that their tools produce valid, reliable predictions and burden individuals’ civil liberties no more than necessary. In evaluating the legality of public health use of algorithms, courts will likely also probe how the output of these tools is used to shape policies and programs. But showing that a model performs well and does not exceedingly burden privacy and other interests are essential preconditions for lawful deployment. ### Privacy law Government intrudes on privacy when it forces people to reveal what …