Moore/Sloan & WRF Innovation in Data Science Postdoctoral Fellowships in the area of

Machine learning, collective variables and non-linear geometric methods in atomistic calculations and molecular dynamics

with Marina Meila (Statistics) and Jim Pfaendtner (Chemical Engineering).

http://escience.washington.edu/get-involved/postdoctoral-fellowships

The goal is to develop and implement methods for optimization, sampling and modeling in large state spaces arising in theoretical chemistry and materials science, using advanced machine learning tools. We are interested in the use of manifold learning methods (also known as non-linear dimension reduction) for problems such as low-dimensional "understandable" representations of samples from large simulations, finding reaction coordinates, saddle points, meta-stable states, as well as for more efficient sampling and optimization in high-dimensional configuration spaces.

A second class of problems will be focused on efficient exploration and optimization in high-dimensional state spaces using active learning.

Qualifications: PhD's in material science, chemistry and related fields who have strong mathematical background AND machine learning/statistics/CS PhD's who are willing to learn about the application are encouraged to apply.

If you are interested in applying for this position, please send the following:

- a short email telling us about your fit with the position and your interests
- CV
- 1-2 significant papers, and slides from talks about these papers if available
- [optionally: the name and email of a person who can give references about you]

to Marina Meila [email protected]. You can do this any time, but we cannot promise to consider your application after December 15, although we will try to do so.