Prospective Students







2012 -13 Seminars

Building an Informatics of Personal Energy Consumption

J. Zico Kolter
Assistant Professor
School of Computer Science at Carnegie Mellon

An important characteristic of many modern energy domains is that they can produce large amounts of data, such as detailed personal consumption information, on an unprecedented scale. The ability to understand this data, to make inferences and predictions about energy information, can play a transformative role in the future of energy systems. In this talk I will discuss how recent advances in machine learning and data analysis can be brought to bear on such problems, and now these problems can themselves motivate new statistical methods. In particular, I will highlight new algorithmic work on energy disaggregation, the task of taking an aggregate power signal and decomposing it into separate devices. This ability helps us understand how energy is consumed in a building, and studies have shown that just presenting this information to users can often lead to large energy savings. I will also discuss work on city-level energy analysis, and how this can inform both customers and cities about the relative energy consumption between homes.