Learning

A system is said to learn if it is capable of acquiring new knowledge from its environment. Learning may also enable the ability to perform new tasks without having to be redesigned or reprogrammed, especially when accompanied by generalization. Learning is most readily accomplished in a system that supports symbolic abstraction, though such a property is not exclusive (reinforcement strategies, for example, do not necessarily require symbolic representation). This type of learning is separated from the acquisition of knowledge through direct programming by the designer, which is referred to throughout this document as the Ability to Add New Knowledge.

Use the following list to see an evaluation of any given architecture along the dimension of ability to learn:

Subsumption Architecture
Gat (Heterogeneous Asynchronous Architecture)
Theo (Plan-then-compile Architecture)
Planning and Learning Architecture (Prodigy)
Modular-Integrated Architecture (ICARUS)
Adaptive Intelligent Systems (AIS)
Basic Integrated Agent (Homer)
Meta-Reasoning Architecture (MAX)
Problem Space Architecture (Soar)
Situated and Planned Action (Teton)
Decision-Theoretic (RALPH-MEA)
Entropy Reduction Engine

Press UP to go to the list of capabilities.

Press HOME to go to the Table of Contents.