At the forefront of modern proliferation detection
diagram showing growing global data streams and traditional intelligence data

ADAPD: Advanced Data Analytics for Proliferation Detection

Nuclear nonproliferation is of paramount concern to the Department of Energy (DOE), the National Nuclear Security Administration (NNSA), and other government agencies. Lawrence Livermore National Laboratory’s (LLNL’s) national security mission includes developing scientific and technological solutions to address the evolving landscape of proliferation threats. This means monitoring and detecting weapons of mass destruction as well as preventing the spread and availability of related materials and infrastructure.

The multi-institutional Advanced Data Analytics for Proliferation Detection (ADAPD, pronounced “adapt”) project aims to make a tangible difference in this crucial mission space through early detection of low-profile proliferation activity that may be small, inaccessible, or buried in background activities. The NNSA’s Office of Defense Nuclear Nonproliferation (DNN) funds ADAPD’s research and development efforts, which draw upon the fruits of another LLNL project.

Proliferation detection relies primarily on the interpretation of quantitative, physics-based observables and the single-modality analysis of these data streams. Although these direct observables can provide high-confidence indicators to characterize known proliferation activities, they cannot support the collection, analysis, and interpretation of observables at the requisite volume, scope of coverage, or time scales needed for early, low-profile detection.

Computing’s Jim Brase, who serves as ADAPD’s venture manager, explains, “We need to look at all potential observables, then put them together in one detection scheme. To do so requires innovative analysis of multimodal data’s cumulative effects. Data science will ultimately enable early detection of low-profile proliferation via techniques more powerful than currently available methods.”

Therefore, ADAPD combines direct and indirect observables—including the technical, business, and human processes that enable and support proliferation activities, such as uranium isotope separation—to deliver a global-scale, real-time capability to detect, locate, and characterize low-profile proliferation.

Launched in 2018 and led by LLNL, the project brings together four other DOE laboratories: Los Alamos, Sandia, Oak Ridge, and Pacific Northwest. Brase notes, “Each lab provides ADAPD with different expertise and data science applications. We’re merging these skills and data sets into a coherent program that tackles a broad problem.” Computing’s Eddy Banks is the project’s principal investigator. Along with LLNL staff from the Engineering, Physics, and Global Security directorates, the ADAPD team includes Computing employees Rafael Rivera Soto, Ted Stirm, and Brian Van Essen.

According to Banks, ADAPD’s roadmap centers on state-of-the-art predictive modeling and analytics techniques. He says, “Our multi-year plan will close existing proliferation detection capability gaps and address the principal components of proliferation analytics through three main research and development areas.”

These areas are (1) predictive models for multiple proliferation observables that integrate new types of data-driven models with traditional physics models and the knowledge of subject matter experts; (2) multi-phenomenology detection to infer the state of hidden proliferation processes; and (3) transfer of models and detection capabilities to new environments.

To implement this strategy, ADAPD relies on Computing’s technical expertise in physics-informed machine learning, multimodal data integration, and nuclear weapons development processes. “New developments in data science, machine learning, and computation provide approaches and tools that are new to this domain,” states Banks. Brase adds, “We will also bring Livermore Computing’s high performance systems to bear on the large-scale machine learning and simulation tasks needed to solve this problem.”

Besides subject-matter expertise and computer power, ADAPD requires data. The amount of accessible data relevant to detecting and characterizing nuclear proliferation activities around the world has grown exponentially over the past decade. Banks explains, “No analytics program can make progress without access to relevant data with well-understood backgrounds and ground truth. Existing proliferation testbeds and experimental activities provide this foundation.”

ADAPD researchers will partner with programs across DOE, NNSA, and DNN to compile data sets that capture the complexity and scale of proliferation activities. The availability of multiple related testbeds provides an opportunity to validate models against different and unknown environments as well as to test validation methods designed to operate in the absence of ground truth.

Indeed, these testbeds’ value cannot be overstated. “Observational data are often unlabeled,” states Banks. “We need to apply ground truth to understand context and interpret patterns, such as seismic patterns caused by construction equipment or another source.” Then, through iterative improvements, computer simulations help reduce noise, produce clean data, and can reconcile sparse or hidden data. The ADAPD team is also developing rigorous uncertainty analyses that, together with modern techniques accounting for model uncertainty, will provide valuable information about model transferability.

Looking ahead, the ADAPD team plans to develop methods that address geospatially distributed and potentially adversarial scenarios in which limited data access and data obfuscation play significant roles. In such scenarios, uncertainty quantification becomes key, and performance of the resulting capabilities will be measured through coordinated NNSA experiments to quantitatively evaluate specific low-profile proliferation scenarios.

“Computing and data science are at the core of the ADAPD effort, opening the door to new approaches to proliferation detection,” summarizes Brase. As such, relationships with industry and universities are becoming increasingly important by providing the collaboration with access to emerging data science advances.

Furthermore, Banks says, “University partnerships have the potential to engage students in the importance of nonproliferation, ensuring that the next generation is equipped to adapt to future proliferation threats.”