Artificial Intelligence
Artificial Intelligence
Applying machine learning
to science and security challenges
Applying machine learning
to science and security challenges
Over the past decade, artificial intelligence (AI) has experienced a renaissance. AI enables machines to learn and make decisions without being explicitly programmed.
AI has enabled a new generation of applications, opening the door to breakthroughs in many aspects of daily life. From situational awareness to threat detection, online signals to system assurance, PNNL is advancing the frontiers of scientific research and national security by applying AI to scientific problems.
Starting with the right environment
For machine learning models, domain-specific knowledge can enhance domain-agnostic data in terms of accuracy, interpretability, and defensibility. PNNL’s AI research has been applied across a variety of domain areas from national security, to the electric grid and Earth systems. Leveraging a deep expertise in the power grid domain, PNNL’s DeepGrid open-source platform uses deep reinforcement learning to help power system operators create more robust emergency control protocols—the safety net of our electric grid. As part of the Physics-Informed Learning Machines for Multiscale and Multiphysics Problems (PhILMs) Center, PNNL is developing physics-informed machine learning techniques.
Building stronger AI systems
PNNL takes a holistic approach to research focused on assuring the safety, security, interpretability, explainability, and general robustness of artificial-intelligence-enabled systems deployed in the real world. This research includes understanding and mitigating system failures caused by design and development flaws, as well as the malicious activities of adversaries.
Revealing the reasoning behind deep learning-based decisions is a critical component of assuring safety, security, and robustness. This reasoning allows PNNL to assess complex systems from the perspective of digital and physical system security, and well as from development and operational perspectives.
Forecasting real-world events
PNNL’s research in content intelligence focuses on the development of novel AI models to explain and predict social systems and behaviors related national security challenges in the human domain. Our expertise in descriptive, predictive, and prescriptive analytics ranges from disinformation detection and attribution to forecasting real-world events such as influenza outbreaks and cryptocurrency price. PNNL’s interactive tools like CrossCheck, ESTEEM and ErrFilter not only ensure we develop robust and generalizable AI models, but also advance understanding and effective reasoning about extreme volumes of dynamic, multilingual, and diverse real-world data.
Integrating across missions
Data engineering is foundational to data science, focusing on information flow from data sources to application. Combining this capability—including expertise in data architectures and pipelines, data collection, and validation—with artificial intelligence enables cross-functional teams to provide optimal solutions to critical mission spaces.
PNNL is teaming with Sandia National Laboratories and Georgia Institute of Technology on the Center for ARtificial Intelligence-focused ARchitectures and Algorithms (ARIAA). The ARIAA team is exploring applying artificial intelligence to address the U.S. Department of Energy’s mission needs in areas, such as cybersecurity and graph analytics. Further, PNNL researchers integrate current approaches for scientific high-performance computing, deep learning, and graph analytics computing paradigms into a converged, coherent computing capability to accelerate scientific discovery.
Gaining great insights from small datasets
While almost all research on few-shot learning is done exclusively on images, PNNL researchers have shown success in other data types, including text, audio, and video. This capability has greatly expanded our AI capabilities beyond traditional, publicly available image datasets and allows researchers to quickly build machine learning models using small amounts of user-classified training examples. Sharkzor, for example, combines human interaction with machine learning techniques to allow classification using just five to ten images—far less than the hundreds or thousands needed for traditional deep learning.
PNNL’s artificial intelligence and machine learning methods and software packages are making a difference in operational environments across the U.S. government and throughout the private sector.