Machine learning of inductive bias
Machine learning of inductive bias
A general theory of discrimination learning
Production system models of learning and development
Mind children: the future of robot and human intelligence
Mind children: the future of robot and human intelligence
Correcting and extending domain knowledge using outside guidance
Proceedings of the seventh international conference (1990) on Machine learning
Learning by experimentation: acquiring and refining problem-solving heuristics
Readings in knowledge acquisition and learning
Learning Search Control Knowledge: An Explanation-Based Approach
Learning Search Control Knowledge: An Explanation-Based Approach
Learning from the environment based on percepts and actions
Learning from the environment based on percepts and actions
Integrated Reactive Soccer Agents
RoboCup-98: Robot Soccer World Cup II
Purposeful Behavior in Robot Soccer Team Play
RoboCup-99: Robot Soccer World Cup III
| Hi-index | 0.00 |
This paper describes an integrated architecture called LIVE designed for learning from the environment. The task that LIVE faces is to solve problems in a new environment using its actions and senses. In order to be able to adapt to different environments, LIVE's actions and senses are defined as "innate" properties and their linkage (i.e., actions and their consequences) is unknown before entering a new environment. LIVE must coordinate a variety of intelligent activities in order to learn from the environment. These include problem solving, exploration, knowledge creation, knowledge revision, experimentation, and discovery. The paper gives an overview of the system and discusses its strengths and weaknesses along several dimensions in comparison with other similar architectures.