Science – Page 22

Self-Driving Cars Can Solve Traffic – By Acting in Their Own Self-Interest

10.11.2025 by qfx

Within a simulated environment of shared control, a learning system navigates the inherent unpredictability of autonomy, demonstrating that even the most sophisticated policies are ultimately grown from, and constrained by, the complexities of the world they inhabit.

New research shows that autonomous vehicles, when trained with artificial intelligence, can improve traffic flow by prioritizing their own goals.

Mapping Market Movements

10.11.2025 by qfx

Local estimations of Gaussian curvature reveal intermittent, regime-like bursts across diverse shapes—from benchmark models to real-world data—suggesting emergent patterns arise not from overarching design, but from localized geometric rules.

New research explores how the geometry of complex spaces can reveal hidden patterns in financial data.

Beyond ImageNet: Training AI to Spot Industrial Defects

10.11.2025 by qfx

Pretrained anomaly representations consistently elevate performance across diverse anomaly detection methods and network architectures, demonstrably surpassing the efficacy of original features as indicated by consistently superior results.

A new pretraining strategy focuses on learning feature representations specifically designed for the challenges of industrial anomaly detection.

When Agents Learn to Trade: Modeling Markets with Diverse Behaviors

10.11.2025 by qfx

A multi-agent reinforcement learning approach models financial markets by assigning unique traits to each agent, then governing their behavior with a shared policy optimized—through optimal transport calibration against real market data—to satisfy individual preferences within a limit order book environment.

New research shows that simulating financial markets with agents who learn and have unique preferences can recreate realistic trading patterns.

Predicting Pasture: Deep Learning Forecasts Irish Grass Growth

10.11.2025 by qfx

Ryegrass, subjected to the predictable rhythms of the months, demonstrates a growth pattern destined to plateau, a temporary flourishing inevitably constrained by inherent limitations.

A new approach to forecasting perennial ryegrass growth leverages the power of deep learning for optimized grassland management.

Mapping the Universe: AI Accelerates Dark Energy Research

10.11.2025 by qfx

The CosmicANNEstimator framework attempts to reconcile observational data with theoretical models through neural networks, yet the very validation process—reliant on Markov Chain Monte Carlo methods—acknowledges the inherent limitations of any framework when confronting the unknown.

A new artificial intelligence method streamlines the process of determining the properties of dark energy using astronomical observations.