We are seeking outstanding Postdoctoral Researchers and PhD Students to develop new machine vision and learning algorithms for cutting-edge neuroscience research. In particular, we are looking for computer scientists with expertise in machine vision and learning interested in both developing new algorithms as well as robust, usable systems that will impact the field of neuroscience.
We are primarily a machine vision and learning lab, developing new technologies for neuroscience research. We are focused on the problem of jointly learning the vocabulary of animal behavior and its implementation in the nervous system, in particular developing machine vision and learning algorithms toward this goal. To do this, we are:
- Combining optogenetic/thermogenetic techniques to manipulate the activity of neurons and new machine vision techniques for extracting the behavioral effects of these manipulations from video and data mining techniques for discovering the underlying structure.
- Jointly learning behavior-anatomy structure from video of animals behaving and video of the neuronal activity (measured via calcium imaging).
- The machine vision and learning systems we develop toward these goals are general-purpose, and used by biologists across a variety of disciplines.
Relevant fields of machine vision and learning include:
- Learning-based pose estimation for high-resolution videos of animals behaving.
- Fully and weakly supervised activity recognition.
- Multi-view clustering/structure discovery.
- Semi-supervised discovery of structure using a "human-in-the-loop" learning paradigm.
The successful candidates for these positions will have:
- The creativity to develop new algorithms and learning paradigms for using machine vision and learning for scientific discovery.
- Practical knowledge of the current state-of-the-art in machine vision and learning.
- The commitment and dedication to develop robust, working systems.
- Interest in applications of computer science to biology, and the new discoveries they enable.
- Strong programming expertise in MATLAB and C, C++, CUDA, Java, or Python.
The application of machine vision and learning to large neuroscience video data sets is an emerging field with great potential impact. New technologies such as optogenetics, calcium imaging, and advances in microscopy have enabled the collection of huge image data sets containing detailed information about the structure and function of the nervous system. Because of the scope and complexity of these data sets, machine vision and learning are of vital importance in extracting scientific understanding. The importance of this field of study has recently been highlighted in Obama's BRAIN Initiative Report (in particular, Section II-4).
HHMI's Janelia Research Campus is a pioneering research center near Washington, D.C., where scientists from many disciplines, from computer science to physics to neuroscience, develop and use emerging and innovative technologies to pursue neuroscience's most challenging problems. Established in 2006, Janelia was modeled after institutes like Bell Labs, with small groups collaborating on high-risk, innovative, big science. All research is internally funded by the Howard Hughes Medical Institute, and salaries and benefits are highly competitive. Postdoc positions at Janelia are renewable one-year appointments. For information about Janelia, please visit: http://www.janelia.org/about-us.
Applicants should email a CV and a cover letter summarizing their research experience and interests to Kristin Branson at firstname.lastname@example.org.