Architectures for Continuously Learning Agents
Abstract
In recent years, progress in artificial intelligence has led to increasingly ambitious efforts to buildintegrated intelligent agents, from robotic agents that perceive and affect the physical world, tosoftware agents that perceive and affect their cyber-world. Despite significant progress in thecomponent parts of these agents (e.g., progress in computer vision, language processing, etc.), akey open question remains: what kind of software architecture is needed to integrate thesecomponents into a continuously self-improving intelligent agent? This is an increasinglyimportant question, as embedded intelligent systems in continuous operation are becomingincreasingly widespread in commercial and military systems. Such continuously operatingsystems, with sensors and effectors that perceive and act on their environment, are exposed to acontinuous stream of data that in many cases could be used for automatic self improvement, ifwe understood how to architect these systems appropriately. Beyond its practical importance,the question of how to architect continuously learning agents is also at the core of the scientificunderstanding of intelligence.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- May 02, 2017
- Source ID
- FA95501710218
Entities
People
- Tom M. Mitchell
Organizations
- Air Force Office of Scientific Research
- Massachusetts Institute of Technology
- United States Air Force