Modeling bottom-up learning from activity in Soar

Olivier L. Georgeon, Frank E. Ritter, Steven R. Haynes

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

This paper presents an implementation of bottom-up learning in a cognitive model. The paper relates the learning mechanism, its implementation in a Soar agent, and an experiment where the agent learns to solve an example task. The agent first learns primary schemas (low-level patterns of behavior) and then secondary schemas (patterns of primary schemas). This implementation draws from theories of hierarchical sequence learning and constructivist learning. It constitutes a first implementation of these theories in Soar, and it shows that Soar can be used to model bottom-up learning, using the weighted preference mechanism available in Soar 9. This study also shows the importance of modeling the agent's activity traces to help the modeler develop the agent. This approach offers a way to represent how autonomous agents create declarative knowledge from and about their interaction with their environment, teammates or opponents.

Original languageEnglish (US)
Title of host publication18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009
Pages65-72
Number of pages8
StatePublished - Dec 1 2009
Event18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009 - Sundance, UT, United States
Duration: Mar 30 2009Apr 2 2009

Publication series

Name18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009

Other

Other18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009
CountryUnited States
CitySundance, UT
Period3/30/094/2/09

Fingerprint

Bottom-up
Autonomous agents
Schema
Modeling
Cognitive Models
Autonomous Agents
Experiments
Trace
Learning
Interaction
Experiment

All Science Journal Classification (ASJC) codes

  • Modeling and Simulation

Cite this

Georgeon, O. L., Ritter, F. E., & Haynes, S. R. (2009). Modeling bottom-up learning from activity in Soar. In 18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009 (pp. 65-72). (18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009).
Georgeon, Olivier L. ; Ritter, Frank E. ; Haynes, Steven R. / Modeling bottom-up learning from activity in Soar. 18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009. 2009. pp. 65-72 (18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009).
@inproceedings{a86cdb94d61049c1bb5f79163c0ea2e2,
title = "Modeling bottom-up learning from activity in Soar",
abstract = "This paper presents an implementation of bottom-up learning in a cognitive model. The paper relates the learning mechanism, its implementation in a Soar agent, and an experiment where the agent learns to solve an example task. The agent first learns primary schemas (low-level patterns of behavior) and then secondary schemas (patterns of primary schemas). This implementation draws from theories of hierarchical sequence learning and constructivist learning. It constitutes a first implementation of these theories in Soar, and it shows that Soar can be used to model bottom-up learning, using the weighted preference mechanism available in Soar 9. This study also shows the importance of modeling the agent's activity traces to help the modeler develop the agent. This approach offers a way to represent how autonomous agents create declarative knowledge from and about their interaction with their environment, teammates or opponents.",
author = "Georgeon, {Olivier L.} and Ritter, {Frank E.} and Haynes, {Steven R.}",
year = "2009",
month = "12",
day = "1",
language = "English (US)",
isbn = "9781615672769",
series = "18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009",
pages = "65--72",
booktitle = "18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009",

}

Georgeon, OL, Ritter, FE & Haynes, SR 2009, Modeling bottom-up learning from activity in Soar. in 18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009. 18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009, pp. 65-72, 18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009, Sundance, UT, United States, 3/30/09.

Modeling bottom-up learning from activity in Soar. / Georgeon, Olivier L.; Ritter, Frank E.; Haynes, Steven R.

18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009. 2009. p. 65-72 (18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Modeling bottom-up learning from activity in Soar

AU - Georgeon, Olivier L.

AU - Ritter, Frank E.

AU - Haynes, Steven R.

PY - 2009/12/1

Y1 - 2009/12/1

N2 - This paper presents an implementation of bottom-up learning in a cognitive model. The paper relates the learning mechanism, its implementation in a Soar agent, and an experiment where the agent learns to solve an example task. The agent first learns primary schemas (low-level patterns of behavior) and then secondary schemas (patterns of primary schemas). This implementation draws from theories of hierarchical sequence learning and constructivist learning. It constitutes a first implementation of these theories in Soar, and it shows that Soar can be used to model bottom-up learning, using the weighted preference mechanism available in Soar 9. This study also shows the importance of modeling the agent's activity traces to help the modeler develop the agent. This approach offers a way to represent how autonomous agents create declarative knowledge from and about their interaction with their environment, teammates or opponents.

AB - This paper presents an implementation of bottom-up learning in a cognitive model. The paper relates the learning mechanism, its implementation in a Soar agent, and an experiment where the agent learns to solve an example task. The agent first learns primary schemas (low-level patterns of behavior) and then secondary schemas (patterns of primary schemas). This implementation draws from theories of hierarchical sequence learning and constructivist learning. It constitutes a first implementation of these theories in Soar, and it shows that Soar can be used to model bottom-up learning, using the weighted preference mechanism available in Soar 9. This study also shows the importance of modeling the agent's activity traces to help the modeler develop the agent. This approach offers a way to represent how autonomous agents create declarative knowledge from and about their interaction with their environment, teammates or opponents.

UR - http://www.scopus.com/inward/record.url?scp=84865345527&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865345527&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84865345527

SN - 9781615672769

T3 - 18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009

SP - 65

EP - 72

BT - 18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009

ER -

Georgeon OL, Ritter FE, Haynes SR. Modeling bottom-up learning from activity in Soar. In 18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009. 2009. p. 65-72. (18th Annual Conference on Behavior Representation in Modeling and Simulation 2009, BRiMS 2009).