Modeling Bottom-Up Learning from Activity in Soar
نویسندگان
چکیده
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.
منابع مشابه
Bottom-Up Recognition Learning: A Compilation-Based Model of Limited-Lookahead Learning
When faced with a novel problem, people can sometimes decide what to do by imagining alternative sequences of actions and then taking the sequence that solves the problem. In many problems, however, various constraints, such as working memory capacity, limit the amount of internal lookahead that people can do. This paper describes Bottom-Up Recognition Learning (BURL), a model of limited-lookah...
متن کاملModeling Single-Task Performance Improvement Using EPIC-Soar
We present a task-independent learning procedure which produces gradual performance improvement through experience on a perceptual-motor ask. It utilizes a data structure that represents a strategy for performing the task. This learning procedure is a direct result of the constraints that occur when realistic perceptual and motor mechanisms (EPIC) are mated to a learning cognitive modeling arch...
متن کاملEFH-Soar: Modeling Education in Highly Interactive Microworlds
The goal of our work is to produce a process model account of education in microworlds based on Soar, a theory of cognition and learning. In the context of a microworld that supports the exploration of qualitative electrostatics, we present operational models of both skilled and student interaction. In addition, we describe an episodic memory mechanism encoded in the student model that gives in...
متن کاملKnowledge Level Learning in Soar
In this article we demonstrate how knowledge level learning can be performed within the Soar architecture. That is, we demonstrate how Soar can acquire new knowledge that is not deductively implied by its existing knowledge. This demonstration employs Soar's chunking mechanism — a mechanism which acquires new productions from goalbased experience — as its only learning mechanism. Chunking has p...
متن کاملCognitive Modeling for Computer Animation: A Comparative Review
Cognitive modeling is a provocative new paradigm that paves the way towards intelligent graphical characters by providing them with logic and reasoning skills. Cognitively empowered self-animating characters will see in the near future a widespread use in the interactive game, multimedia, virtual reality and production animation industries. This review covers three recently-published papers fro...
متن کامل