Interpretable Climate Change Modeling With Progressive Cascade Networks
This work addresses the need for more interpretable models in climate science, though it appears incremental in its approach.
The paper tackles the problem of interpretability in deep learning for climate modeling by introducing a method that builds models incrementally from linear to more complex forms, applied to mapping global temperature and precipitation to years to investigate climate change patterns.
Typical deep learning approaches to modeling high-dimensional data often result in complex models that do not easily reveal a new understanding of the data. Research in the deep learning field is very actively pursuing new methods to interpret deep neural networks and to reduce their complexity. An approach is described here that starts with linear models and incrementally adds complexity only as supported by the data. An application is shown in which models that map global temperature and precipitation to years are trained to investigate patterns associated with changes in climate.