 In his Explorer series talk, NCAR scientist Dr. David John Gagnier discusses seeing the atmosphere through machine learning. Machine learning is essentially a special kind of computer program. The human shows the machine learning model a lot of different examples of what the machine learning model should see and then what it should expect to happen. In the case of weather prediction, we would show the machine learning model the equivalent of a weather map and then tell it, a tornado happened today or a tornado didn't happen today. And in doing so, it eventually learns the underlying patterns that separate these two situations. From this talk, I want people to take away three things. First, that machine learning is a powerful tool that can be used for improving weather predictions. I'm going to show people how these algorithms work, especially for helping us understand our climate system better. And finally, I'm going to talk about what are the challenges associated with machine learning.