Soar

Theory

Soar embodies multiple hypotheses about the computational structures underlying general intelligence, many of which are shared with other cognitive architectures, including ACT-R, which was created by John R. Anderson, and LIDA, which was created by Stan Franklin. Recently, the emphasis on Soar has been on general AI (functionality and efficiency), whereas the emphasis on ACT-R has always been on cognitive modeling (detailed modeling of human cognition).

The original theory of cognition underlying Soar is the Problem Space Hypothesis, which is described in Allen Newell's book, Unified Theories of Cognition.

Robotics

Many different robotic applications have been built using Soar since the original Robo-Soar was implemented in 1991 for controlling a Puma robot arm. Later, a generalized version of Merle-Soar (Dispatcher-Soar) was used to demonstrate a symbolic, constraint propagation approach in learning to improve schedules and to define task-independent knowledge metrics of architecture-specific learning -- knowledge efficiency, knowledge utility, and knowledge effectiveness.

Music

Melody-Soar demonstrated how the Soar architecture could explain and demonstrate creativity in simple melody generation using hierarchies of problems spaces that parallel the hierarchical structure of melody, allowing unique melodies to be generated from preferences of existing styles (e.g., Bach).

See also

  • Cognitive Architecture
  • ACT-R

References

Bibliography

  • Laird, 2012 The Soar Cognitive Architecture
  • Lehman, Laird, and Rosenbloom, 2006 A Gentle Introduction to Soar: 2006 update
  • Rosenbloom, Laird, and Newell, 1993 The Soar Papers: Readings on Integrated Intelligence , Information Sciences Institute
  • Soar Homepage on University of Michigan
  • Soar: Frequently Asked Questions List
  • Soar Tech Homepage
  • Paul Rosenbloom