Towards collaborative and adversarial learning: a case study in robotic soccer
โ Scribed by PETER STONE; MANUELA VELOSO
- Publisher
- Elsevier Science
- Year
- 1998
- Tongue
- English
- Weight
- 330 KB
- Volume
- 48
- Category
- Article
- ISSN
- 1071-5819
No coin nor oath required. For personal study only.
โฆ Synopsis
Soccer is a rich domain for the study of multiagent learning issues. Not only must the players learn lower-level skills, but they must also learn to work together and to adapt to the behaviors of different opponents. We are using a robotic soccer system to study these different types of multiagent learning: low-level skills, collaborative and adversarial. Here we describe in detail our experimental framework. We present a learned, robust, low-level behavior that is necessitated by the multiagent nature of the domain, viz. shooting a moving ball. We then discuss the issues that arise as we extend the learning scenario to require collaborative and adversarial learning.
๐ SIMILAR VOLUMES
## Abstract A method of learning dating to the 1950s is founded on the conviction that people learn best from their experience, but only if they can reflect critically on that experience. Action learning is a way to learn while doing; it balances the action of work with shared reflection among part
## Abstract Information systems (IS) projects involving multiple organizations are very common today. Knowledge integration in such projects is a complex task of integrating diverse knowledge bases across organizations that may possess distinct strategic goals and even conflicting interests. Prior
## Abstract Research has revealed that fetuses can learn from events in their environment. The most convincing evidence for fetal learning is habituation to vibroacoustic stimulation (VAS) in human fetuses and classical conditioning in rat fetuses. However, these two research areas have been indepe