一项新研究揭示了不同模型训练方法的优缺点。
Journal of Geophysical Research: Machine Learning and Computation
Posted inResearch Spotlights
Comparing Machine Learning Models of Raindrop Formation
The simplest model, based on polynomials, yields the best performance.
Posted inResearch Spotlights
AI Improves Earthquake Detection
A new study shows the pros and cons of different model training methods.
Posted inEditors' Highlights
Collinearity is Not Always a Problem in Machine Learning
Collinearity is not always a showstopper for statistical machine learning (at least not for self-organizing maps).
Posted inEditors' Vox
Bridging the Gap: Transforming Reliable Climate Data into Climate Policy
A new special collection welcomes research that bridges the gap between rigorous Essential Climate Variable (ECV) monitoring, AI analytics, and climate policy.
Posted inResearch Spotlights
AI Sheds Light on Hard-to-Study Ocean Currents
The Maluku Strait is a key predictor of conditions in the Indonesian Throughflow, modeling shows.
Posted inResearch Spotlights
A Step Toward AI Modeling of the Whole Earth System
Coupling an AI-driven model of the atmosphere with a model of the ocean could help scientists create highly efficient emulations of the entire Earth system.
Posted inResearch Spotlights
Unveiling What’s Under the Hood in AI Weather Models
Artificial intelligence models have improved weather forecasting, but their inner workings are largely opaque. A new approach could make their predictions more interpretable by scientists.
Posted inEditors' Vox
Advances in Ecological Forecasting
AGU and ESA invite contributions to a cross-society special collection on ecological forecasting across ecosystems and scales.
Posted inEditors' Highlights
Machine Learning Provides a New Perspective of Low-level Clouds
Low-level clouds over the oceans, extensively studied for their role in climate change, are re-examined from a new perspective that applies machine learning to radar observations.