Science – Page 21 – Minority Mindset

Currency Havens in an Age of Uncertainty

11.11.2025 by qfx

New research reveals how emerging market currencies react to global economic shocks, identifying potential safe-haven assets.

Predicting Food Donations with Adaptive AI

10.11.2025 by qfx

Even the most carefully constructed embodied force field (EFB) will ultimately succumb to the unpredictable pressures of real-world interaction.

A new framework leverages reinforcement learning to improve the accuracy of food donation forecasts, even as demand and supply shift.

Can Digital Minds Fall for Fake News?

10.11.2025 by qfx

Agent personas, when subjected to simulated misinformation, exhibit varied reactions, illuminating the inherent fragility of systems as they navigate deceptive information landscapes.

New research shows that simulated people powered by artificial intelligence react to misinformation much like humans do.

The Hidden Carbon Footprint of AI

10.11.2025 by qfx

Sustainability isn't achieved through incremental steps, but through a tiered system of rigorous checkpoints—G-TRACE indicators assessing carbon output, lifecycle impact, and regional energy factors—that either propel systems forward or demand further optimization before advancing toward true climate stewardship.

As generative AI scales, the energy demands of running these models are quickly surpassing those of training, creating a significant and often overlooked climate impact.

How Vulnerable Is Your AI to Hidden Attacks?

10.11.2025 by qfx

Each model possesses a unique adversarial subspace, yet these subspaces inevitably overlap, demonstrating that even disparate systems share vulnerabilities which will, in time, become points of systemic failure.

New research offers a way to measure the risk of adversarial attacks that transfer from unseen sources, even without knowing the attacker’s methods.

Decoding Cyber Threats with AI’s Explanations

10.11.2025 by qfx

Retrieval-Augmented Generation (RAG) systems, when applied to a Blockchain-based Cyber Threat Intelligence (CTI) dataset, demonstrate a capacity to match answers, though the inherent complexity suggests even elegant solutions will eventually succumb to the realities of production-level data.

New research explores how advanced AI can not only identify cyber threats, but also clearly explain why.

Predicting Road Risk with Probabilistic Deep Learning

10.11.2025 by qfx

The assessment of crash risk along the San Antonio River Walk revealed that a probabilistic model—by prioritizing nuanced spatial coherence and superior recall of historical fatal crashes—generated a more realistic safety evaluation than a baseline model prone to abrupt fluctuations and missed critical locations.

A new approach uses the full power of probability to more accurately assess crash likelihood on roadways.

Predicting Heart Risk in Diabetes: A New Era of Accuracy

10.11.2025 by qfx

Machine learning and deep learning techniques are significantly improving the prediction of cardiovascular disease in diabetic patients.

Predicting Carbon Prices: A New Approach to Forecasting Emissions Markets

10.11.2025 by qfx

Multiple forecasting models were evaluated, revealing varying degrees of accuracy in predicting carbon price fluctuations.

A novel forecasting framework combining signal processing and deep learning dramatically improves the accuracy of EU carbon price predictions.

Beyond the Algorithm: Making Credit Decisions Transparent

10.11.2025 by qfx

The neural network, despite its complexity, demonstrates a predictable, instance-by-instance rationale, suggesting that even sophisticated models ultimately rely on discernible, albeit potentially opaque, logic when making predictions.

New research details a practical framework for evaluating how easily we can understand the reasoning behind AI-powered credit risk assessments.