Computational Sustainability
I heard a really nice talk at WiML (Workshop for Women in Machine Learning) this year. One of the speakers suggested finding people to work with in your favorite science and seeing what types of machine learning problems they had. I really liked this advice/idea. One of my friends does wildlife and natural resource conservation research. Over the years we have had several conversations about some of the things she is doing in research. Several of the problems she has talked about would be very interesting machine learning problems. I also ran into Tom Dietterich at NIPS who mentioned a project about fire intervention in forests that he is working on. While I was searching around for postdoc opportunities I found out about Cornell's Institute for Computational Sustainability that is just starting http://www.computational-sustainability.org/ . I think that this is a really interesting effort and will keep this type of research in mind as I look for interesting research problems in the future.