Human and Information Interaction

Research concentrates on understanding and exploiting interactions between information and humans. It involves complex mixed-initiative processes of information acquisition, processing and comprehension. Aspects of this research complement efforts in the Human Sciences Campaign, with the delineation being that research in the Information Sciences Campaign places greater emphasis on information structure, dynamics, phenomena and properties as well as information’s relationship to machine learning and computational reasoning.

Discovery Mechanisms for Engendering Creative Decision Making (ALC)

The objective of this research is to investigate new techniques for engendering creativity to improve decision making. In a data-driven context, the intelligence phase of decision making involves exploring relevant data, derived analysis and formulating alternatives and options based on this process.  This leads to very structured approaches to alternative assessment and evaluation; resulting in the obfuscation of creative or non-typical solutions.  Humans operating in environments where data is pervasive have to make data-driven decisions faster, given ever-accumulating amounts of information and problems that require increasing amounts of creativity.  Research in this topic seek to address challenges of constrained creativity through the merger of large data and the development and use of computer models that exhibit, generate and evaluate creativity in the intelligence phase of decision making. Research areas will include recommender technologies and other mathematically-grounded methodologies that reason over humans’ profiles, intent, and goals in order to explore and discover information to spur decision-relevant ideas and “out-of-the-box” alternatives.

Principal Investigator:

Adrienne Raglin, adrienne.j.raglin.civ@mail.mil, (301) 394-2475

Abductive Reasoning Under Uncertainty for Accelerated Decision Making (ALC)

The objective of this research is to investigate new techniques for alignment and explanation of reasoning in artificial intelligence.  Commanders’ must respond to the complex, diverse, and dynamic nature of current and future operational environment.   To exploit asymmetric views and accelerate decision making in Army operations, Soldiers will rely on agents and enabling technologies such as complex systems-of-systems that integrate intelligent sensor networks and autonomous systems. These systems-of-systems will be driven by machine learning enabled artificial intelligence (ML/AI) algorithms, forming teams with human warfighters, where both must act as one unit. Research in this topic seeks to address the challenges of creating computationally-derived explanations of uncertainty quantification and artificial intelligence reasoning. Research areas should include probabilistic abductive reasoning and other mathematically-grounded methodologies that reason over multiple, and likely conflicting, explanations that result from multiple ML/AI models such that uncertainty in subsequent decision making is reduced.

Principal Investigator:

Adrienne Raglin, adrienne.j.raglin.civ@mail.mil, (301) 394-2475