Big steps forward are synonymous with new ideas – a thought that merits mindfulness if we aspire to train students to do more than merely train machines.
machine learning & AI
Posted inNews
Machine Learning Can Help Decode Alien Skies—Up to a Point
Astronomers are testing the tools that might help them keep up with the upcoming storm of exoplanet atmosphere data.
Posted inNews
Teaching Machines to Detect Climate Extremes
Artificial intelligence can be used to analyze massive amounts of data from climate simulations, but more training data are needed.
Posted inResearch Spotlights
Improving Atmospheric Forecasts with Machine Learning
An efficient, low-resolution machine learning model can usefully predict the global atmospheric state as much as 3 days out.
Posted inNews
How Machine Learning Redraws the Map of Ocean Ecosystems
Using an unsupervised learning algorithm, scientists can create new maps of ecosystem provinces in the ocean, opening the possibility of sharper data collection and monitoring.
Posted inFeatures
Visualizing Science: How Color Determines What We See
Color plays a major role in the analysis and communication of scientific information. New tools are helping to improve how color can be applied more accurately and effectively to data.
Posted inScience Updates
Creating Data Tool Kits That Everyone Can Use
Earth scientists outline challenges to making the growing wealth of available data more accessible and to using data services for interdisciplinary research and applications.
Posted inNews
Are Cosmic Rays a Key to Forecasting Volcanic Eruptions?
A combination of relativistic particles and artificial intelligence may provide a new way to forecast when a volcano could erupt.
Posted inEditors' Highlights
A New Global Map of Seafloor Fluid Expulsion Anomalies
The first open-source database of SEAfloor FLuid Expulsion Anomalies (SEAFLEASs) at a global scale reveals their distribution and physical parameters.
Posted inResearch Spotlights
Machine Learning Improves Weather and Climate Models
New research evaluates the performance of generative adversarial networks for stochastic parameterizations.