Can News Sentiment Predict Stock Swings?
A new study explores how analyzing the emotional tone of financial news using advanced AI can improve the accuracy of stock price forecasting.
Science
Decoding Energy Price Signals: A Clearer Path to Forecasting
![The model learns a feature shape [latex]f(x)[/latex] that captures heating demand across multiple price zones in response to upwards-moving frequency regulation requests, revealing how demand profiles are reshaped by market dynamics.](https://arxiv.org/html/2602.00049v1/x1.png)
A new study demonstrates that transparent forecasting models can match the accuracy of complex algorithms in predicting short-term energy market prices.
Fragile Explanations: How Model Uncertainty Undermines AI Trust
New research reveals that the reliability of counterfactual explanations – a key tool for understanding AI decisions – is surprisingly vulnerable to the inherent uncertainties within machine learning models.
Trading with Physics: AI Learns from Newton’s Laws to Optimize Portfolios
![A reinforcement learning framework optimizes portfolio weights-represented as [latex]\mathbf{w}_{t}[/latex]-based on state and price relative information, updating portfolio value [latex]P_{t}[/latex] while incorporating transaction costs and, crucially, embedding a physics-informed loss function [latex]L_{\text{phys}}[/latex] derived from Newton’s second law [latex]F=ma[/latex]-specifically, the discrepancy between predicted and actual acceleration-to guide the optimization process.](https://arxiv.org/html/2602.01388v1/Flowchart_of_PINN.png)
A new approach leverages principles of Newtonian dynamics within deep reinforcement learning algorithms to enhance financial trading strategies.
Predicting Power Prices: Deep Learning in Australia’s Energy Market
A new review assesses the effectiveness of deep learning techniques for forecasting electricity prices in the Australian National Electricity Market, a crucial task for efficient grid management and renewable energy integration.
Can Algorithms Beat the Market? A New Benchmark for Prediction Trading
Researchers have introduced a comprehensive framework for rigorously testing and comparing trading strategies in prediction markets, paving the way for more robust algorithmic and AI-powered approaches.
Beyond Task Completion: Can Language Models Truly Investigate?
New research challenges large language models to move beyond following instructions and demonstrate genuine data exploration skills.
The Evolving Market: Simulating Finance as an Ecosystem
A new framework models financial markets not as static systems, but as dynamic ecosystems where diverse strategies compete and adapt, offering insights into systemic risk and policy effectiveness.
Can AI Find the Next Hot Stock?
New research shows that generative artificial intelligence, when properly informed, can identify previously unseen factors driving stock performance.
Building Agents That Learn to Use Tools, From Simulated Experience to Real-World Tasks
Researchers have developed a new framework for training large language models to interact with tools and solve complex problems through a combination of self-generated training data and reinforcement learning.