Researchers have developed a method that combines statistical techniques to reveal underlying drivers in financial time series data, offering improved insights for risk management and portfolio construction.
Researchers are using evolutionary algorithms to discover novel activation functions that enhance the ability of neural networks to generalize to unseen data.
Researchers have developed a novel framework that tackles the challenges of identifying anomalies in dynamic, imbalanced graph data.
New research leverages advanced graph neural networks to uncover hidden relationships between congressional trading and stock market performance.
New research tackles the problem of ‘hallucination’ – the tendency of AI models to invent facts – when applied to complex financial data and analysis.
A new architecture leveraging Bernstein polynomials as activation functions promises improved gradient flow, representational power, and potential for significant network compression.
New research reveals that privacy-preserving machine learning techniques can significantly degrade performance on datasets where some classes are far more common than others.
A new framework explores how generative AI and online Q&A forums can move beyond competition and forge a mutually beneficial relationship.
A new AI framework uses reinforcement learning to drastically speed up and improve the process of identifying optimal locations for affordable housing developments.
A novel learned optimization framework improves the stability and convergence of solutions for ill-posed inverse problems, offering a significant advancement in fields like brain imaging.