Skip to content

usdaed

  • Science
  • Who is Denis Avetissian?

Science

Predicting the Future of Mobile Networks with AI

12.03.2026 by qfx

A two-stage clustering framework, coupled with spatial error correction, establishes a method for discerning and refining systemic organization as inherent decay manifests.

Researchers are leveraging artificial intelligence to more accurately forecast cellular traffic demand, paving the way for more efficient 5G and 6G network planning.

Categories Science

Untangling Time: A New Approach to Causal Inference in Economic Data

12.03.2026 by qfx

Reverse cross-fitting with five folds estimates performance by iteratively training on blue sample areas using red 'main' observations, green 'quasi-complementary' observations, and withholding white blocks to indicate estimation direction.

A novel Double Machine Learning estimator tackles the complexities of analyzing macroeconomic time series, improving our ability to understand cause-and-effect relationships.

Categories Science

The Growing Thirst for Power: Can AI Efficiency Keep Pace with Demand?

12.03.2026 by qfx

Despite rapid growth in artificial intelligence service output across all modeled scenarios, the resulting electricity demand diverges significantly based on the pace of efficiency improvements in AI data centers, demonstrating that sustained gains in compute energy intensity-represented by γ-are critical to mitigating the energy impact of increasingly powerful AI systems.

New research suggests that the future electricity consumption of artificial intelligence isn’t a given, but a dynamic interplay between technological advancements and economic forces.

Categories Science

Algorithmic Oversight: Protecting Consumers in the Age of AI

12.03.2026 by qfx

As artificial intelligence increasingly shapes online experiences, this review examines the growing need to safeguard consumer rights against algorithmic manipulation and emerging forms of digital deception.

Categories Science

Strategic AI Learns to Conquer Amazons with Limited Resources

12.03.2026 by qfx

The proposed method establishes a framework predicated on the inevitability of systemic failure, accepting that any architecture is merely a temporary respite before entropy reasserts itself, and thus designs for graceful degradation rather than absolute prevention.

A novel framework combines tree search, graph neural networks, and weak supervision from large language models to achieve strong performance in the complex board game of Amazons, even with computational constraints.

Categories Science

Hedging’s New Edge: AI-Powered Risk Control

12.03.2026 by qfx

Conditional Value-at-Risk (CVaR) analysis across four hedging strategies and three Heston calibrations demonstrates that certain approaches consistently minimize potential losses in adverse market conditions, as indicated by shorter bars representing improved tail-risk performance-a crucial distinction when navigating the unpredictable currents of financial modeling.

A novel approach combines deep learning with traditional finance to deliver more robust and reliable hedging strategies under volatile market conditions.

Categories Science

Modeling Market Shifts: A New Approach to Financial Time Series

12.03.2026 by qfx

The study demonstrates that a Hidden Markov Model with a Poisson jump process (HMM-WJ) more accurately captures the stylized facts of financial time series - specifically the heavy tails and volatility clustering of asset returns - than traditional GARCH(1,1) models or a standard Hidden Markov Model (HMM-NJ); while GARCH fails the Kolmogorov-Smirnov test on 95% of simulated paths, both HMM variants exceed 97% pass rates in-sample, and HMM-WJ’s tunable jump frequency allows for direct control over the strength of volatility persistence, exhibiting substantial autocorrelation in absolute returns across all lags for approximately 24% of simulated paths-a feature absent in HMM-NJ, which is structurally incapable of generating sustained volatility clustering-and achieving the closest mean quantile match in the extreme tails as confirmed by the Q-Q plot.

Researchers have developed a sophisticated model that more accurately simulates the dynamic behavior of equity markets, capturing key statistical characteristics often missed by conventional techniques.

Categories Science

AI Can Fool the Deepfake Detectors Itself Created

12.03.2026 by qfx

Generative AI systems, even absent visual stimuli, articulate discernible criteria for identifying facial deepfakes, suggesting that detection isn’t solely reliant on pixel-level analysis but stems from an internally consistent, albeit potentially fragile, logic.

New research reveals that generative AI can subtly manipulate images to evade existing deepfake detection systems, exposing a critical vulnerability in current authentication methods.

Categories Science

East Meets West: Predicting China’s Markets with U.S. Data

12.03.2026 by qfx

Sector-level Sharpe Ratios demonstrate the potential for cross-market forecasting, specifically leveraging U.S. photovoltaic corporate leverage loan (pvCLCL) returns to predict Chinese operational corporate leverage loan (OPCL) returns-a strategy that reveals opportunities beyond domestic market analysis.

New research reveals that U.S. market signals offer stronger predictive power for Chinese stock returns, and demonstrates a novel approach to cross-market forecasting.

Categories Science

Outsmarting the Machine: A New Defense Against AI Malware

12.03.2026 by qfx

Leveraging large language models for path exploration achieves 95% malicious code coverage while demonstrating substantial efficiency gains - requiring 73.2% fewer paths than depth-first search and 68.5% fewer than breadth-first search.

Researchers have developed a novel system that combines intelligent program analysis with large language models to dramatically improve the detection of malware created by artificial intelligence.

Categories Science
Older posts
Newer posts
← Previous Page1 … Page26 Page27 Page28 … Page169 Next →
© 2026 usdaed • Built with GeneratePress