Example: stock market

The Soar User’s Manual Version 9.6

The Soar User s ManualVersion E. Laird, Clare Bates Congdon,Mazin Assanie, Nate Derbinsky and Joseph XuAdditional contributions by:Mitchell Bloch, Karen J. Coulter, Steven Jones,Aaron Mininger, Preeti Ramaraj and Bryan StearnsDivision of Computer Science and EngineeringUniversity of MichiganDraft of: November 2, 2017 Errors may be reported to John E. Laird 1998 - 2017, The Regents of the University of MichiganDevelopment of earlier versions of this Manual were supported under contract N00014-92-K-2015 from the Advanced Systems Technology Office of the Advanced Research ProjectsAgency and the Naval Research Laboratory, and contract N66001-95-C-6013 from the Ad-vanced Systems Technology Office of the Advanced Research Projects Agency and the NavalCommand and Ocean Surveillance Center, RDT&E Using this Manual .

Soar has been developed to be an architecture for constructing general intelligent systems. It has been in use since 1983, and has evolved through many di erent versions. This manual documents the most current of these: version 9.6.0. Our goals for Soar include that it ultimately be an architecture that can: be used to build systems that work ...

Tags:

  Rosa

Information

Domain:

Source:

Link to this page:

Please notify us if you found a problem with this document:

Other abuse

Advertisement

Transcription of The Soar User’s Manual Version 9.6

1 The Soar User s ManualVersion E. Laird, Clare Bates Congdon,Mazin Assanie, Nate Derbinsky and Joseph XuAdditional contributions by:Mitchell Bloch, Karen J. Coulter, Steven Jones,Aaron Mininger, Preeti Ramaraj and Bryan StearnsDivision of Computer Science and EngineeringUniversity of MichiganDraft of: November 2, 2017 Errors may be reported to John E. Laird 1998 - 2017, The Regents of the University of MichiganDevelopment of earlier versions of this Manual were supported under contract N00014-92-K-2015 from the Advanced Systems Technology Office of the Advanced Research ProjectsAgency and the Naval Research Laboratory, and contract N66001-95-C-6013 from the Ad-vanced Systems Technology Office of the Advanced Research Projects Agency and the NavalCommand and Ocean Surveillance Center, RDT&E Using this Manual .

2 Contacting the Soar Group .. Different Platforms and Operating Systems ..42 The Soar An Overview of Soar .. of Procedural Knowledge in Soar .. Functions in Soar .. Example Task: The Blocks-World .. of States, Operators, and Goals .. candidate operators .. candidate operators: Preferences .. a single operator: Decision .. the operator .. inferences about the state .. Problem Spaces .. Working memory: The Current Situation .. Production Memory:Long-term Procedural Knowledge .. structure of a production .. roles of productions .. Actions and Persistence .. Preference Memory: Selection Knowledge.

3 Semantics .. preferences are evaluated to decide an operator .. Soar s Execution Cycle: Without Substates .. Input and Output .. Impasses and Substates .. Types .. New States .. : Support for results .. : Learning Procedural Knowledge .. calculation of o-support .. of Substates: Impasse Resolution .. s Cycle: With Substates .. of Substates: The Goal Dependency Set ..373 The Syntax of Soar Working Memory .. preferences in working memory .. Memory as a Graph .. Memory Activation .. Preference Memory .. Production Memory .. Names .. string (optional) .. type (optional) .. (optional).

4 Condition side of productions (or LHS) .. action side of productions (or RHS) .. for production syntax .. Impasses in Working Memory and in Productions .. in working memory .. for impasses in productions .. Soar I/O: Input and Output in Soar .. of Soar I/O .. and output in working memory .. and output in production memory ..904 Procedural Knowledge Chunking .. Explanation-based Chunking .. Overview of the EBC Algorithm .. Five Main Components of Explanation-Based Chunking .. What EBC Does Prior to the Learning Episode .. Assignment and Propagation .. Operator Selection Knowledge Tracking .. What EBC Does During the Learning Episode.

5 The Complete Set of Results .. and the Three Types of Analysis Performed .. Formation .. Subtleties of EBC .. Between Chunks and Justifications .. Inhibition .. Based on Chunks .. Chunks and Justifications .. and Correctness of Learned Rules .. and Over-generalization .. Results and Rule Repair .. Operator Selection Knowledge .. Over Operators Selected Probabilistically .. Collapsed Negative Reasoning .. Problem-Solving That Doesn t Test The Superstate .. Disjunctive Context Conflation .. Generalizing knowledge retrieved fromsemantic or episodic memory .. Learning from Instruction.

6 Determining Which OSK Preferences are Relevant .. Generalizing Knowledge From Mathand Other Right-Hand Side Functions .. Situations in which a Chunk is Not Learned .. Usage .. of thechunkcommand .. Procedural Learning .. What Your Agent Learns .. What Was Learned .. Explaining Learned Procedural Knowledge .. Visualizing the Explanation .. 1285 Reinforcement RL Rules .. Reward Representation .. Updating RL Rule Values .. in Rule Coverage .. and Substates .. Traces .. ( ) .. Automatic Generation of RL Rules .. gp Command .. Templates.

7 1416 Semantic Working Memory Structure .. Knowledge Representation .. Long-Term Identifiers with Soar .. Storing Semantic Knowledge .. command .. command .. Storage .. Location .. Retrieving Semantic Knowledge .. Retrievals .. Retrievals .. with Depth .. Performance .. queries .. Tweaking .. 1537 Episodic Working Memory Structure .. Episodic Storage .. Contents .. Location .. Retrieving Episodes .. Retrievals .. Non-Cue-Based Retrieval .. Non-Cue-Based Retrieval .. Meta-Data .. Performance.

8 Tweaking .. 1628 Spatial Visual The scene graph .. Scene Graph Edit Language .. Commands .. and extractonce .. Filters .. lists .. List .. Writing new filters .. subclasses .. Node Filters .. Command line interface .. 1819 The Soar User Basic Commands for Running Soar .. Procedural Memory Commands .. Short-term Memory Commands .. Learning .. Long-term Declarative Memory .. Other Debugging Commands .. File System I/O Commands .. System .. 292 Index295 Summary of Soar Aliases, Variables, and Functions301viCONTENTSList of Soar is continually trying to select and apply operators.

9 The initial state and goal of the blocks-world task.. The initial state of the blocks world as working memory objects .. The WM state in blocks world after the first operator is selected .. Six proposed blocks world operators .. The blocks-world problem space .. An abstract view of production memory .. The preference resolution process .. A detailed illustration of Soar s decision cycle.. A simplified Version of the Soar algorithm.. A simplified illustration of a subgoal stack.. Simplified Representation of the context dependencies .. The Dependency Set in Soar.. A semantic net illustration of four objects in working memory.

10 An example production from the example blocks-world task.. An example portion of the input link for the blocks-world task.. An example portion of the output link for the blocks-world task.. A Soar chunk vs. an explanation-based chunk .. A comparison of a working memory trace and an explanation trace .. A visualization of an explanation trace .. An explanation trace of two simple rules that matched in a substate .. An explanation trace after identity analysis .. The five main components of explanation-based chunking .. The seven stages of rule formation .. A colored visualization of an explanation trace.


Related search queries