mlsanityData to 7 Oct 2026

Knowledge-Integrated Informed AI for National Security

arXiv 2202.03188 ↗First posted 4 Feb 2022cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

The state of artificial intelligence technology has a rich history that dates back decades and includes two fall-outs before the explosive resurgence of today, which is credited largely to data-driven techniques. While AI technology has and continues to become increasingly mainstream with impact across domains and industries, it's not without several drawbacks, weaknesses, and potential to cause undesired effects. AI techniques are numerous with many approaches and variants, but they can be classified simply based on the degree of knowledge they capture and how much data they require; two broad categories emerge as prominent across AI to date: (1) techniques that are primarily, and often solely, data-driven while leveraging little to no knowledge and (2) techniques that primarily leverage knowledge and depend less on data. Now, a third category is starting to emerge that leverages both data and knowledge, that some refer to as "informed AI." This third category can be a game changer within the national security domain where there is ample scientific and domain-specific knowledge that stands ready to be leveraged, and where purely data-driven AI can lead to serious unwanted consequences. This report shares findings from a thorough exploration of AI approaches that exploit data as well as principled and/or practical knowledge, which we refer to as "knowledge-integrated informed AI." Specifically, we review illuminating examples of knowledge integrated in deep learning and reinforcement learning pipelines, taking note of the performance gains they provide. We also discuss an apparent trade space across variants of knowledge-integrated informed AI, along with observed and prominent issues that suggest worthwhile future research directions. Most importantly, this report suggests how the advantages of knowledge-integrated informed AI stand to benefit the national security domain.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 40 of its 92 references that are in the map. First posted on arXiv 4 Feb 2022.

Builds on

40 of 92 references.

References · 40

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2021On the Opportunities and Risks of Foundation Modelscited by 3,961
  5. 2017Deep reinforcement learning from human preferencescited by 3,846
  6. 2018Neural Ordinary Differential Equationscited by 3,733
  7. 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,806
  8. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,327
  9. 2019The frontier of simulation-based inferencecited by 490
  10. 2020Universal Differential Equations for Scientific Machine Learningcited by 345
  11. 2017A Semantic Loss Function for Deep Learning with Symbolic Knowledgecited by 267
  12. 2020Neurosymbolic AI: The 3rd Wavecited by 253
28 more
  1. 2019Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learningcited by 205
  2. 2014Robots that can adapt like animalscited by 203
  3. 2019Informed Machine Learning -- A Taxonomy and Survey of Integrating Knowledge into Learning Systemscited by 203
  4. 2020The Next Decade in AI: Four Steps Towards Robust Artificial Intelligencecited by 194
  5. 2020Reward Machines: Exploiting Reward Function Structure in Reinforcement Learningcited by 183
  6. 2019Explainable Machine Learning for Scientific Insights and Discoveriescited by 165
  7. 2020Integrating Scientific Knowledge with Machine Learning for Engineering and Environmental Systemscited by 160
  8. 2017Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial Differential Equationscited by 158
  9. 2019Neural-Symbolic Computing: An Effective Methodology for Principled Integration of Machine Learning and Reasoningcited by 149
  10. 2019Hamiltonian Generative Networkscited by 136
  11. 2017Deep TAMER: Interactive Agent Shaping in High-Dimensional State Spacescited by 122
  12. 2016A Differentiable Physics Engine for Deep Learning in Roboticscited by 107
  13. 2017Deep Learning for Physical Processes: Incorporating Prior Scientific Knowledgecited by 95
  14. 2020Graph Neural Networks Meet Neural-Symbolic Computing: A Survey and Perspectivecited by 93
  15. 2016Exploiting Cyclic Symmetry in Convolutional Neural Networkscited by 91
  16. 2016Label-Free Supervision of Neural Networks with Physics and Domain Knowledgecited by 74
  17. 2016The More You Know: Using Knowledge Graphs for Image Classificationcited by 63
  18. 2020Backpropagation through Signal Temporal Logic Specifications: Infusing Logical Structure into Gradient-Based Methodscited by 51
  19. 2018SPIGAN: Privileged Adversarial Learning from Simulationcited by 47
  20. 2018Inference in Probabilistic Graphical Models by Graph Neural Networkscited by 38
  21. 2019MPC-Net: A First Principles Guided Policy Searchcited by 28
  22. 2017Beating Atari with Natural Language Guided Reinforcement Learningcited by 24
  23. 2018Physics-informed deep generative modelscited by 16
  24. 2020Improving Deep Learning Models via Constraint-Based Domain Knowledge: a Brief Surveycited by 16
  25. 2018From Deep to Physics-Informed Learning of Turbulence: Diagnosticscited by 14
  26. 2018Hybrid Knowledge Routed Modules for Large-scale Object Detectioncited by 14
  27. 2020Universal Battery Performance and Degradation Model for Electric Aircraftcited by 8
  28. 2018Symmetry constrained machine learningcited by 3

Built on by

Not cited yet.