A University of Washington model uses the past 40 years of weather events to make predictions. Not only do the predictions provide a high level of accuracy, the process of making the predictions the uses 7,000 times less computer power compared with current technology. The importance of the accelerated speed is that it will enable weather forecasting centers to run many models with slightly different starting conditions far faster. This should help improve the predictions around extreme weather events, like hurricanes.
The AI was trained by undertaking similar repeated steps from one forecast to the next, in order to build up process knowledge using Microsoft technology. To accelerate the training the Earth was modelled as a cube, and then flattened. Upon this different weather patterns were imposed. This approach enabled the AI to learn rapidly.
According to lead researcher Dr. Jonathan Weyn: “Machine learning is essentially doing a glorified version of pattern recognition.”
The academic explains further, adding: “It sees a typical pattern, recognizes how it usually evolves and decides what to do based on the examples it has seen in the past 40 years of data.”
The research appears in the journal Journal of Advances in Modeling Earth Systems. The research paper is titled “Improving Data‐Driven Global Weather Prediction Using Deep Convolutional Neural Networks on a Cubed Sphere.”
In related news, Finnish researchers have reported on a new machine-human cooperation to provide ecological information. The new form of artificial intelligence can explain the relation between the area and age of an island and the number of species it hosts. This system provides new insights about the way ecosystems or species behave in space and time.
This helps to answer questions like:
Why do some species exist in some regions and not in others?
Why do some regions have more species than others?
The University of Helsinki development appears in the journal Frontiers in Ecology and Evolution. The research paper is titled “Automated Discovery of Relationships, Models, and Principles in Ecology.”
