Prodigy

• Philosophy and Methodological Assumptions
• Description of the Architecture
  • Learning Modules

• Style of Control
• Forward and Backward Chaining
• Modular Organization
• Symbolic World Model (Global State)
• Glass Box Approach
• Declarative Representations
• Global Representation and Uniform Access to Knowledge
• Homogeneous (Uniform) Knowledge Representation
• First-Order Logic Representation
• STRIPS-like Representation
• Deliberative Learning
• Generalization
• Monotonic Learning
• The Utility Function
• Means-End Analysis Technique
• Deliberation/Operation Speed
• Casual Commitment
• List of Agent Properties for All Architectures

• Multi-Method Learning
  • Explanation-Based Learning
  • Static Analysis for Control Rule Learning
  • Analogical Learning
  • Abstraction Learning
  • Refining and Correcting Domain Knowledge
  • Learning from an Expert
• Planning
• Problem Solving
• Meta-Reasoning
• Self-Reflection
• Prediction
• Focused Behavior and Processing
• List of Capabilities for All Architectures

• Static Environments
• Environmental Consistency
• Simulated Environments
• Knowledge-Rich
• Complete Knowledge
• Learning Cost
• List of Environmental Considerations for All Architectures

References

Current Location: Prodigy - Top