Fighting Fraud, Together and Privately
A new approach to detecting payment fraud leverages the power of multiple institutions without sharing sensitive customer data.
Science
Beyond Personalization: Crafting Intent-Driven Financial Marketing
A new framework leverages the power of artificial intelligence to deliver hyper-relevant and compliant marketing messages in the financial sector.
Smarter Oilfields: AI Spots Hidden Losses and Theft
A new approach leverages the power of network analysis and deep learning to identify energy waste, fraudulent activity, and operational bottlenecks in complex oil and gas production systems.
Riding the Wave: LSTM Networks Boost Equity Trend Forecasting
A new approach leveraging Long Short-Term Memory networks and differencing techniques dramatically improves the performance of trend-following strategies in financial markets.
AI Takes the Reins: Automating Factor Investing for Alpha
A new framework uses artificial intelligence to autonomously discover and implement profitable investment strategies, potentially reshaping quantitative portfolio management.
Beyond Algorithms: The Rise of Agentic Finance
The increasing deployment of AI agents is fundamentally reshaping financial markets, moving beyond traditional model-based automation.
Unraveling Time Series: From Moments to Meaning
![The method distills complex time-series data into class-wise global explanations by first generating local explanations with [latex]LOMATCE[/latex], then strategically selecting a representative subset of instances-prioritizing coverage of influential, merged event clusters-to aggregate parameterised event primitives.](https://arxiv.org/html/2603.13065v1/x1.png)
A new model-agnostic approach bridges the gap between local explanations and comprehensive, class-wide insights in time series classification.
Economic Survival of the Fittest: Forecasting with Network Data
A new agent-based model demonstrates surprisingly accurate macroeconomic forecasts using only input-output tables and principles of Darwinian selection.
Cleaning the Cosmic Signal: How Diverse Data Makes Neural Networks More Reliable
![The study assesses the accuracy and precision-quantified by [latex]Eqs. 13 \text{ and } 14[/latex]-of various architectures in modeling cosmic microwave background polarization, specifically E-modes and B-modes across a multipole range of [latex]50 < \ell < 260[/latex], and demonstrates how performance is influenced by the foreground model employed during training.](https://arxiv.org/html/2603.12364v1/x15.png)
New research shows that training neural networks on increasingly complex simulated data significantly improves their ability to remove foreground noise from cosmic microwave background polarization maps.
Decoding the Market with Knowledge and Reasoning
A new framework leverages the power of knowledge graphs and artificial intelligence to not only predict stock movements but also explain why those movements are likely to happen.