Lawrence Livermore National Laboratory



Reinforcement Learning Postdoctoral Researcher

Location:  Livermore, CA
Category:  Post Docs
Organization:  Engineering
Posting Requirement:  External Posting
Job ID: 106135
Job Code: Post-Dr Research Staff 1 (PDS.1)
Date Posted: September 24 2019

Share this Job

Apply Now

Apply For This Job

Join us and make YOUR mark on the World!

Come join Lawrence Livermore National Laboratory (LLNL) where we apply science and technology to make the world a safer place; now one of 2020 Best Places to Work by Glassdoor!

We have multiple openings for a Postdoctoral Researcher to conduct basic and applied research in Reinforcement Learning for optimal sequential decision-making under uncertainty, on real-world problems in healthcare, cyber security, and national security applications. These positions are in the Computational Engineering Division (CED), within the Engineering Directorate.

Essential Duties
- Conduct research in reinforcement learning to enable development of new state-of-the-art algorithms for Laboratory problem domains.
- Design, implement, and analyze techniques in reinforcement learning.
- Explore techniques for controlling simulations using approaches such as deep reinforcement learning, bandit optimization, evolutionary algorithms, model-based methods, and stochastic control.
- Contribute to and actively participate in the conception, design, and execution of research to address defined problems.
- Pursue independent (but complementary) research interests and interact with a broad spectrum of scientists internally and externally to the Laboratory.
- Collaborate with others in a multidisciplinary team environment to accomplish research goals.
- Publish research results in peer-reviewed scientific or technical journals and present results at external conferences seminars and/or technical meetings.
- Perform other duties as assigned.

Qualifications
- Recent PhD in Computer Science, Computational Engineering, Applied Statistics, Applied Mathematics, Operation Research, or related field.
- Knowledge and experience in reinforcement learning, active learning, or stochastic control algorithms.
- Experience in one or more of the following machine learning areas: deep learning, unsupervised feature learning, multimodal learning, and probabilistic graphical models.
- Experience in the broad application of two or more higher-level programming languages such as Python, Java, Matlab, R or C/C++.
- Experience with one or more deep learning libraries such as TensorFlow, Keras, Caffe or Theano, and experience with one or more deep reinforcement learning libraries such as rllab, keras-rl or OpenAI Gym.
- Experience developing independent research projects, including publication of peer-reviewed literature.
- Proficient verbal and written communication skills to collaborate effectively in a team environment and present and explain technical information.
- Initiative and interpersonal skills and ability to work in a collaborative, multidisciplinary team environment.

Desired Qualifications
- Strong math background and experience with mathematical formulations of complex systems. Domain knowledge in biological sciences or cyber security.
- Experience with a variety of deep reinforcement learning algorithms including experience with variational Bayesian methods, nonparametric Bayesian methods, and multi-agent systems.
- Experience with utilizing simulation to model and analyze complex systems and experience with parallel computing and/or GPU computing.

Pre-Employment Drug Test: External applicant(s) selected for this position will be required to pass a post-offer, pre-employment drug test.  This includes testing for use of marijuana as Federal Law applies to us as a Federal Contractor.

Security Clearance:  This position requires either no security clearance, or a Department of Energy (DOE) L-level or Q-level clearance depending on the particular assignment.

If you are selected and a security clearance is required, we will initiate a Federal background investigation to determine if you meet eligibility requirements for access to classified information or matter. In addition, all L or Q cleared employees are subject to random drug testing.  L and Q-level clearances require U.S. citizenship.  If you hold multiple citizenships (U.S. and another country), you may be required to renounce your non-U.S. citizenship before a DOE L or Q clearance will be processed/granted.

If no security clearance is required, but your assignment is longer than 179 days cumulatively within a calendar year, you must go through the Personal Identity Verification process.  This process includes completing an online background investigation form and receiving approval of the background check.  (This process does not apply to foreign nationals.)

Note: This is a one year Postdoctoral appointment with the possibility of extension to a maximum of three years. Eligible candidates are recent PhDs within five years of the month of the degree award at time of hire date.

About Us

Lawrence Livermore National Laboratory (LLNL), located in the San Francisco Bay Area (East Bay), is a premier applied science laboratory that is part of the National Nuclear Security Administration (NNSA) within the Department of Energy (DOE). LLNL's mission is strengthening national security by developing and applying cutting-edge science, technology, and engineering that respond with vision, quality, integrity, and technical excellence to scientific issues of national importance. The Laboratory has a current annual budget of about $2.1 billion, employing approximately 6,800 employees.

LLNL is an affirmative action/equal opportunity employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, marital status, national origin, ancestry, sex, sexual orientation, gender identity, disability, medical condition, protected veteran status, age, citizenship, or any other characteristic protected by law.