Search for: All records

This work expands on previous advancements in genetic fingerprint spoofing via the DeepMasterPrints and introduces Diversity and Novelty MasterPrints. This system uses quality diversity evolutionary algorithms to generate dictionaries of artificial prints with a focus on increasing coverage of users from the dataset. The Diversity MasterPrints focus on generating solution prints that match with users not covered by previously found prints, and the Novelty MasterPrints explicitly search for prints with more that are farther in user space than previous prints. Our multi-print search methodologies outperform the singular DeepMasterPrints in both coverage and generalization while maintaining quality of the fingerprint image output.

With projected temperature increases and extreme events due to climate change for many regions of the world, characterizing the impacts of these emerging hazards on water distribution systems is necessary to identify and prioritize adaptation strategies for ensuring reliability. To aid decision-making, new insights are needed into how water distribution system reliability to climate-driven heat will change, and the proactive maintenance strategies available to combat failures. To this end, we present the model Perses, a framework that joins a water distribution network hydraulic solver with reliability models of physical assets or components to estimate temperature increase-driven failures and resulting service outages in the long term. A theoretical case study is developed using Phoenix, Arizona temperature profiles, a city with extreme temperatures and a rapidly expanding infrastructure. By end-of-century under hotter futures there are projected to be 1%–5% more pump failures, 2%–5% more PVC pipe failures, and 3%–7% more iron pipe failures (RCP 4.5–8.5) than a baseline historical temperature profile. Service outages, which constitute inadequate pressure for domestic and commercial use are projected to increase by 16%–26% above the baseline under maximum temperature conditions. The exceedance of baseline failures, when compounded across a large metro region, reveals potential challenges formore »budgeting, management, and maintenance. An exploration of the mitigation potential of adaptation strategies shows that expedited repair times are capable of offsetting the additional outages from climate change, but will come with a cost.

« less

Misinformation has developed into a critical societal threat that can lead to disastrous societal consequences. Although fact-checking plays a key role in combating misinformation, relatively little research has empirically investigated work practices of professional fact-checkers. To address this gap, we conducted semi-structured interviews with 21 fact-checkers from 19 countries. The participants reported being inundated with information that needs filtering and prioritizing prior to fact-checking. The interviews surfaced a pipeline of practices fragmented across disparate tools that lack integration. Importantly, fact-checkers lack effective mechanisms for disseminating the outcomes of their efforts which prevents their work from fully achieving its potential impact. We found that the largely manual and labor intensive nature of current fact-checking practices is a barrier to scale. We apply these findings to propose a number of suggestions that can improve the effectiveness, efficiency, scale, and reach of fact-checking work and its outcomes.

Motivated by the need to rapidly and equitably upskill workers in response to technological advances and social change, a team funded by the US National Science Foundation Convergence Accelerator program has developed and piloted an application called SkillSync that (i) helps employers identify and communicate the skills their workers require; (ii) connects employers to non-degree college programs that can provide relevant training; (iii) helps those programs align their offerings with employer requirements; and (iv) facilitates the exchange of proposals to offer training. This paper describes SkillSync and explains how Artificial Intelligence (AI) is used to automate skills extraction and align training resources with training requests. This paper then discusses the steps taken to ameliorate possible biases and an intelligent agent that is included in SkillSync to increase transparency and trust in the application. The methods discussed in this paper are applicable to many classes of applications that use AI in training, education, and talent management

Deep generative models have enabled the automated synthesis of high-quality data for diverse applications. However, the most effective generative models are specialized to data from a single domain (e.g., images or text). Real-world applications such as healthcare require multi-modal data from multiple domains (e.g., both images and corresponding text), which are difficult to acquire due to limited availability and privacy concerns and are much harder to synthesize. To tackle this joint synthesis challenge, we propose an End-to-end MultImodal X-ray genERative model (EMIXER) for jointly synthesizing x-ray images and corresponding free-text reports, all conditional on diagnosis labels. EMIXER is an conditional generative adversarial model by 1) generating an image based on a label, 2) encoding the image to a hidden embedding, 3) producing the corresponding text via a hierarchical decoder from the image embedding, and 4) a joint discriminator for assessing both the image and the corresponding text. EMIXER also enables self-supervision to leverage vast amount of unlabeled data. Extensive experiments with real X-ray reports data illustrate how data augmentation using synthesized multimodal samples can improve the performance of a variety of supervised tasks including COVID-19 X-ray classification with very limited samples. The quality of generated images and reports are alsomore »confirmed by radiologists. We quantitatively show that EMIXER generated synthetic datasets can augment X-ray image classification, report generation models to achieve 5.94% and 6.9% improvement on models trained only on real data samples. Taken together, our results highlight the promise of state of generative models to advance clinical machine learning.« less