Wednesday, 25 January 2017: 9:00 AM
310 (Washington State Convention Center )
Detecting extreme events in large datasets is a major challenge in climate science research. Current algorithms for extreme event detection are built upon human expertise in defining events based on subjective thresholds of relevant physical variables. Often, multiple competing methods produce vastly different results on the same dataset. Accurate, objective characterization of extreme events in climate simulations and observational data archives is critical for understanding the trends and potential impacts of such events in a climate change context. This study presents the first application of Deep Learning techniques as alternative methodology for climate extreme event detection. Deep neural networks are able to learn high-level representations of a broad class of patterns from labeled data. We have developed a deep Convolutional Neural Network (CNN) classification system, and we have demonstrated the usefulness of Deep Learning techniques for tackling climate pattern detection problems. Coupled with Bayesian based hyper-parameter optimization scheme, our deep CNN system achieves 89%-99% of accuracy in detecting extreme events including tropical cyclones, atmospheric rivers, and weather fronts.
![](images/flag.png)
![](images/blue_ribbon.png)