Capitalizing on Conflict: The FORR Architecture
نویسنده
چکیده
FORR is a general architecture for learning and problem solving. Its design harnesses multiple, disagreeing perspectives on a domain and learns to make intelligent choices from their conflicting advice. FORR has been implemented in Hoyle, a program that learns to play two-person, perfect information board games. A FORR-based program is modular, robust in the face of error or lack of information, degrades gracefully, and should improve gradually with problem solving experience. This paper describes a theory for skill acquisition, the knowledge base FORR requires, and the way FORR’s design affects learning.
منابع مشابه
Applying FORR to human/multi-robot teams
This extended abstract briefly describes a new methodology for shared decision-making in human/multi-robot teams. We present a new application of FORR, a cognitive architecture that considers the opinions of others when choosing actions. While robots in our system ultimately make their own decisions about their actions, they do so based on collective advice from humans, agents and other robots.
متن کاملFor the Right Reasons: The FORR Architecture for Learning in a Skill Domain
The theme of this article is that knowledge acquisition can be managed as a transition from general expertise to specific expertise. FORR is a general architecture for learning and problem solving that models expertise at a set of related problem classes. This architecture postulates initial broad domain knowledge, and gradually specializes it to simulate expertise in individual problem classes...
متن کاملBuilding a Constraint Solver that Learns
FORR (FOrr the Right Reasons) is a cognitively plausible architecture for the rapid development of expertise (Epstein, 1992). To produce an adaptive, robust problem solver, FORR exploits many techniques observable in human learners. FORR itself is domain independent; a FORR-based application requires a set of domain-specific state representations and heuristics. Hoyle, a FORR-based game player,...
متن کاملCollaboration and Interdependence among Limitedly Rational Agents
FORR is a learning and problem-solving architecture that capitalizes upon synergy among a variety of limitedly rational agents. It takes the position that many reasonable but imperfect decision-making agents can, when they agree upon a course of action, quickly make choices that are good enough and will improve with learning. Some FORR agents react quickly and correctly to a small amount of sen...
متن کاملLearning to Support Constraint Programmers
This paper describes the Adaptive Constraint Engine (ACE), an ambitious ongoing research project to support constraint programmers, both human and machine. The program begins with substantial knowledge about constraint satisfaction. The program harnesses a cognitively oriented architecture—FOr the Right Reasons (FORR) to manage search heuristics and to learn new ones. ACE can transfer what it l...
متن کامل