Machine Learning Applications for Nuclear Security
Bethany Goldblum, Research Engineer
Nuclear Engineering
Closed. This professor is continuing with Spring 2024 apprentices on this project; no new apprentices needed for Fall 2024.
Effective nuclear proliferation detection is hindered by the need to continuously verify the absence of undeclared nuclear materials and nuclear weapons-relevant activities. Multisensor data fusion has the potential to provide an integrated picture of difficult to detect phenomena, where composite signals can be used as proliferation indicators. Recent developments in ultra low power wireless sensor networks, in concert with advances in complexity science and statistical machine learning, offer a path forward for innovation in data analytics for proliferation detection and nuclear material security. This work uses a network of multi-sensor devices deployed at a nuclear reactor and reprocessing facility. The goal of this work is to fuse information from the many sensors and modalities to measure complex events reflective of nuclear proliferation.
The ideal candidate will have software development skills and an interest in working on concepts at the intersection of nuclear science and security policy.
Role: The URAP student will assist in the development of machine learning algorithms in Python for pattern recognition using heterogeneous input data. The student is required to attend and participate in a weekly research group meeting.
Qualifications: Required:
Undergraduate degree in progress; Interest in the intersection of science and nuclear security policy; Creative approach to problem-solving; Software development skills in Python
Desired:
Upper division undergraduate standing; Studies in mathematics, statistics, computer science, physics, or related fields
Hours: 9-11 hrs
Off-Campus Research Site: Remote or in-person options at the Nuclear Science and Security Consortium; 2150 Shattuck Ave, Suite 230; Berkeley, CA
Related website: http://complexity.berkeley.edu/
Related website: http://appliedphysics.nuc.berkeley.edu/