J. Zico Kolter
Stanford University Ph.D.
MIT Postdoc

Sustainable energy issues pose one of the largest challenges facing
society: 84% of the world's energy currently comes from fossil fuels,
raising major issues with climate change, energy security, and the
long-term availability of these sources.  Although energy domains span a
huge range of different areas, a common theme in many modern energy tasks
is the availability of large amounts of data, and the need to learn models,
make inferences, and control the system based upon this data.  These are
problems that require new methods in machine learning, probabilistic
inference, and control, and where such algorithms can have a profound
impact on the energy space.  In this talk I will look at two particular
tasks spanning different extremes of energy consumption and generation and
show how new algorithmic methods can play a pivotal role in each.

First, on the energy consumption side, I will present new techniques for
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 directly lead to large energy
savings.  Unlike previous approaches to this problem, my work considers
models that look jointly at the entire signal and exploit the rich temporal
structure of the data.  The key technical challenge here is the task of
making inferences in these high-dimensional, factorized, temporal models,
and I will present new algorithms I have developed, based upon convex
relaxations of inference, that greatly outperform existing approaches on
this task.  Second, on the energy generation side, I will present work on
maximizing power output for wind turbines in low-wind conditions.  In
particular, I will present a novel policy learning approach, based upon
trust-region optimization, which is able to maximize power using much less
data than existing learning techniques, and which produces 30% more power
than a purely model-based approach on an experimental wind turbine.