2020 AMS Annual Meeting

Preliminary Agenda

Machine learning has become prominent within the atmospheric and environmental sciences over the past years.  Numerical model parameterization, empirical predictive modeling, data post-processing, and many other sub-fields have benefitted from the rapid introduction of machine learning techniques into our community.  It is likely that the importance of machine learning will only continue to expand as computers become more efficient and new methods are developed.  Whether you want to apply machine learning within your own research or gain a better understanding of methods being used by your colleagues, this course will help prepare you for an era where machine learning has an ever-growing impact on our field.

This course is a hands-on hackathon that will allow participants to actively apply machine learning algorithms to a problem in the atmospheric and environmental sciences.  The primary goal of this course is to give attendees experience applying machine learning within a scientific setting.  Teams will build a machine learning model, which will include choosing which type of machine learning model to apply to the data, training the model, and improving the accuracy of the model through parameter tuning.  Attendees will be instructed by active researchers within the atmospheric science/machine learning community.  This course is designed for individuals with previous background in machine learning, either through the introductory AMS course the day prior or through their own research.

Instructors include atmospheric scientists and machine learning experts from various academic institutions and government labs.

Participants are expected to bring their own laptops to be able to participate in the hands-on exercises.  Separate instructions will be sent out to participants regarding additional set-up instructions before the workshop.

For more information, please contact either Ben Toms ([email protected]) or Amanda Burke ([email protected]).