Skip to content

usdaed

  • Science
  • Who is Denis Avetissian?

Science

Can AI Spot the Hidden Hand in Options Markets?

23.12.2025 by qfx

Detection confidence consistently exceeded 79% across three distinct trading patterns-gamma positioning, stock pinning, and 0DTE hedging-with every instance surpassing a 60% mechanical threshold, indicating robust and reliable identification of these strategies within the analyzed data (detection counts: 168, 163, and 188, respectively).

New research shows large language models can infer the subtle forces of dealer hedging, even when stripped of identifying data, suggesting a deeper understanding of market mechanics.

Categories Science

Unmasking Collusion: How Network Science Detects Insider Trading

23.12.2025 by qfx

The network analysis reveals how centrality within a system isn’t simply about volume, but the quality of connection-strong, yellow-hued ties indicating robust relationships, while weaker, purple-tinged links suggest more tenuous affiliations-demonstrating that influence propagates not just through many, but through the <i>right</i> connections.

A new approach leverages the power of network analysis to identify coordinated trading activity among corporate insiders, revealing patterns hidden from traditional surveillance.

Categories Science

Beyond Location: Injecting Spatial Knowledge into Deep Forecasting

23.12.2025 by qfx

Standard self-attention mechanisms, prone to overfitting, learn noisy long-range correlations, while a geostatistical attention approach enforces a smooth, topology-aware prior consistent with an underlying Gaussian Random Field, offering a potential path toward more robust and generalizable models.

A new approach integrates geostatistical principles into transformer networks to improve the accuracy and efficiency of predicting events across space and time.

Categories Science

Smart Grids Get a Brain Boost: AI-Powered Transmission Switching

23.12.2025 by qfx

A dispatch-aware deep neural network learns to optimize transmission switching through a training process designed to anticipate and accommodate systemic failures inherent in any complex network.

Researchers are leveraging deep learning to optimize power grid control, enabling faster and more reliable responses to changing demand and conditions.

Categories Science

Beyond the Algorithm: Islamic Ethics and the Fight Against Deepfakes

23.12.2025 by qfx

As deepfake technology advances, a new ethical framework is needed to address the potential for misinformation and harm, and this paper proposes Islamic principles as a uniquely effective solution.

Categories Science

Seeing Around Corners: Predictive Radar for Safer Seas

22.12.2025 by qfx

Automatic Identification System (AIS) data provides vessel-specific information-such as identity and voyage details-limited to cooperative vessels, whereas X-band radar, as demonstrated by the MOANA dataset, captures a comprehensive scene including shoreline and potentially ambiguous echoes from clutter or navigation aids, highlighting a fundamental difference in the scope and data provided by these two maritime observation technologies.

This review explores how artificial intelligence can anticipate future radar returns, enabling autonomous vessels to navigate more effectively in complex maritime environments.

Categories Science

Predicting the Plug-In Future: How Machine Learning Forecasts EV Charging Demand

22.12.2025 by qfx

The study addresses the critical need for accurate forecasting of electric vehicle (EV) charging loads, a task essential for stabilizing power grids and optimizing energy distribution as EV adoption increases.

Accurately forecasting electric vehicle charging load is crucial for grid stability and efficient energy management as EV adoption accelerates.

Categories Science

Unmasking the Hidden Biases in AI Image Generation

22.12.2025 by qfx

The study reveals that text-to-image models, when prompted with queries flagged by MineTheGap as potentially biased, exhibit limited semantic diversity-generating remarkably similar outputs-while a standard image search retrieves a wider range of visual interpretations for the same prompts, exposing the models’ constrained understanding of nuanced requests.

Researchers have developed a new automated system to expose the subtle prejudices embedded within text-to-image models, revealing how these systems can perpetuate harmful stereotypes.

Categories Science

Mapping the Alzheimer’s Brain: A New Path to Early Diagnosis

22.12.2025 by qfx

Researchers are leveraging advanced network analysis and machine learning to unlock deeper insights into the brain changes associated with Alzheimer’s disease.

Categories Science

Hidden in Plain Sight: Uncovering Illicit Data on Ethereum

22.12.2025 by qfx

A framework establishes a method for integrating data restoration and analysis within the Ethereum blockchain, acknowledging the inevitable decay of systems and prioritizing sustained functionality rather than simply measuring elapsed time, as data persists within the network’s evolving structure-a process defined not by duration but by the medium of the blockchain itself-and allowing for continuous assessment even as the underlying system ages.

Researchers are now able to detect and analyze sensitive and illegal content embedded within Ethereum transactions, revealing a previously unseen risk to blockchain security and privacy.

Categories Science
Older posts
Newer posts
← Previous Page1 … Page125 Page126 Page127 … Page171 Next →
© 2026 usdaed • Built with GeneratePress