Soar capabilities
- Reinforcement Learning
- Chunking
- Look-ahead Subgoaling
- Can learn procedural knowledge
In this category, you'll find a variety of agents developed for a wide range of different tasks and environments. You will find a full description of the agent, its capabilities and problem solving approach as well, as a download link. If the agent requires an environment, a link will be provided. We'd highly encourage you to submit your own agents for use by the greater Soar community. To do so, you can send your zipped up submission with a full description (try to include all of the type of information we include on each download page) to mazina@umich.edu with the subject "Soar Agent Submission".
-
Blocks-World (Look-Ahead with RL)
-
Created by:
Soar
Blocks-World (Look-Ahead with RL)
This agent contains a version of blocks-world that involves one level of problem spaces and look-ahead but with two important extensions. It demonstrates how RL-rules can be learned by chunking and then updated in the future. The advantage over simple lookahead is that it doesn't lock on to the one path found during look-ahead after chunking. It will still do some exploration.
Soar capabilities -
Created by:
-
Blocks-World (Look-Ahead)
-
Created by:
Blocks-World (Look-Ahead)
This agent contains a version of blocks-world that involves one level of problem spaces and look-ahead. It is based on the simple blocks-world agent.
Soar capabilities- Look-ahead subgoaling
- Can learn procedural knowledge (enable with 'learn always')
- None.
- simple.soar
- selection.soar
- The Soar Cognitive Architecture: Chapter 4
- John Laird
- Soar 8,
-
Created by:
-
Blocks-World (Simple)
-
Created by:
Blocks-World (Simple)
A very simple instance of an internally-simulated blocks world with three blocks. Uses two operator evaluation rules to eliminate all search.
Soar capabilities- Basic PSCM functions: State Elaboration, Operator Proposal, Operator Evaluation, Internal Operator Application
- None.
- None.
- The Soar Cognitive Architecture: Chapter 4
- John Laird
- Soar 8,9
-
Created by:
-
Counting Agent
-
Created by:
Counting Agent
Agents that can count in nine different ways.
The real purpose of this agent is to provide various test of Soar's execution:- Long runs
- Lots of working memory
- Various subgoals/impasses and results with chunking
- Lots of chunks/results
- Nasty conditions for chunking
- Should find any memory leaks.
Soar capabilities- Subgoaling
- Chunking
External Environment- None.
- None.
- None.
- John
-
Created by:
-
Eaters (Hello World Rule)
-
Created by:
Eaters (Hello World Rule)
A very simple agent (1 production) that prints out "Hello World" and halts.
Soar capabilities- Basic production syntax
- Printing a debug statement
- Halting agent run from a production
- None.
- The Soar Tutorial: Chapter 2
- John Laird
- Soar 8,9
- Simple text
-
Created by:
-
Eaters (Hello World Operator)
-
Created by:
Eaters (Hello World Operator)
A very simple agent (2 productions) that prints out "Hello World" and halts. This agent differs from the previous hello world agent insofar as proposes an operator and then prints its output in the operator's application.
Soar capabilities- Basic PSCM functions: Operator Proposal, Internal Operator Application
- None.
- The Soar Tutorial: Chapter 2
- John
-
Created by:
-
Episodic Memory Tutorial Agent
-
Created by:
Episodic Memory Tutorial Agent
A very simple agent that shows how structures are stored in episodic memory and then retrieved.
Soar capabilities- Episodic memory
- None.
- None.
- Soar Episodic Memory Tutorial
- John Laird
- Soar 9
- Plain text
-
Created by:
-
Episodic Memory Counting Agent
-
Created by:
Episodic Memory Counting Agent
This agent is a self-validating unit test for episodic memory.
Parameter settings for EpMem are set in firstload. Parameters for the agent (as described below) are set in the action of the initialization application rule.
It first counts to some number (max), triggering episodic storage at each tick (this is mode "recording"). Episodes include both an integer count (S1 ^symbolic.episode) as well as a digit-by-digit symbolic representation (S1 ^symbolic.converted.*).... -
Created by:
-
Left-Right Reinforcement Learning Agent
-
Created by:
Left-Right Reinforcement Learning Agent
A very simple agent that demonstrates reinforcement learning. Agent chooses between left and right and must learn that left is preferred.
Soar capabilities- Reinforcement learning
- None.
- None.
- Soar Reinforcement Learning Tutorial
- John Laird
- Soar 9
- VisualSoar
-
Created by:
-
Semantic Memory Tutorial Agent
-
Created by:
Semantic Memory Tutorial Agent
A very simple agent that shows how structures are stored in semantic memory and then retrieved.
Soar capabilities- Semantic memory
- None.
- None.
- Soar Semantic Memory Tutorial
- John Laird
- Soar 9
- Plain text
-
Created by:
There are no articles in this category.