This paper is the recommended initial reading for a functional overview of Soar, version 9.6. It supplements more in-depth descriptions of Soar, such as the Soar book (Laird, 2012); the Soar Tutorial (Laird, 2017), which provides a step-by-step introduction to programming Soar; and the Soar Manual , which gives complete details of Soar's operation and use.Soar is meant to be a general cognitive architecture (Langley et al., 2009) that provides the fixed computational building blocks for creating AI agents whose cognitive characteristics and capabilities approach those found in humans (Laird, 2012;Newell, 1990). A cognitive architecture is not a single algorithm or method for solving a specific problem; rather, it is the task-independent infrastructure that learns, encodes, and applies an agent's knowledge to produce behavior, making a cognitive architecture a software implementation of a general theory of intelligence. One of the most difficult challenges in cognitive architecture design is to create sufficient structure to support coherent and purposeful behavior, while at the same time providing sufficient flexibility so that an agent can adapt (via learning) to the specifics of its tasks and environment. The structure of Soar is inspired by the human mind and as Allen Newell (Newell, 1990) suggested over 30 years ago, it attempts to embody a unified theory of cognition.Soar was originally developed in the early 1980s as an architecture to support multi-task, multi-method problem solving. Its evolution has led to new modules and capabilities, including reinforcement learning, episodic and semantic memory, and the spatial visual system Laird, 2012). It is a freely available open-source project (https://soar.eecs.umich.edu/). Over the years, a wide range of agents has been developed in Soar, many of which are briefly described in the appendix. These include agents embodied in real-world robots, computer games, and large-scale distributed simulation environments. These agents incorporate combinations of real-time decision-making, planning, natural language understanding, metacognition, theory of mind, mental imagery, and multiple forms of learning. Soar's focus has mainly been on AI agents, but it has been used for detailed modeling of human behavior (Stearns, 2021;Schatz et al., 2022).A key hypothesis of Soar is that there are sufficient regularities above the neural level to capture the functionality of the human mind. Thus, the majority of knowledge representations in Soar are symbol structures, with architecturally maintained numeric metadata biasing the retrieval and learning of those structures. Soar also includes supports non-symbolic reasoning through the spatial visual system, which is an interface between perception and working memory. There is no commitment to a specific underlying implementation level, such as using simulated neurons. Our commitment is to an efficient and portable implementation, which happens to be in C/ C++ (a Java version duplicates the functionality of Soar 9.3). All agents develop...
Abstract. In this article we describe an approach to the construction of a general learning mechanism based on chunking in Soar. Chunking is a learning mechanism that acquires rules from goal-based experience. Soar is a general problem-solving architecture with a rule-based memory. In previous work we have demonstrated how the combination of chunking and Soar could acquire search-control knowledge (strategy acquisition) and operator implementation rules in both search-based puzzle tasks and knowledge-based expert-systems tasks. In this work we examine the anatomy of chunking in Soar and provide a new demonstration of its learning capabilities involving the acquisition and use of macro-operators.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.