面向AGI的代表论、功能论与自然主义智能概念
人工智能
2025-03-11 v1 机器人学
摘要
本文分析了与创建人工通用智能(AGI)相关的基础原则。将智能理解为在先前未知条件下创建新技能以实现目标的能力。为此，智能利用演绎、归纳和归谬等推理方法，以及抽象和分类等其他方法来开发世界模型。这些方法应用于通过感知等方式获得的对世界的间接且不完整的表征，这些表征并不描绘世界而仅对应世界。由于这些限制以及推理的不确定性和偶然性，世界模型是建构主义的。其价值功能性地由其可行性决定，即其实现所需目标的潜在能力。结果，意义被赋予表征，通过赋予其实现目标所需功能来实现。这种代表论和功能论智能观 enables a naturalistic interpretation that does not presuppose mental features, such as intentionality and consciousness, which are regarded as independent of intelligence. Based on a phenomenological analysis, it is shown that AGI can gain a more fundamental access to the world than humans, although it is limited by the No Free Lunch theorems, which require assumptions to be made.
引用
@article{arxiv.2503.07600,
title = {A Representationalist, Functionalist and Naturalistic Conception of Intelligence as a Foundation for AGI},
author = {Rolf Pfister},
journal= {arXiv preprint arXiv:2503.07600},
year = {2025}
}