DeepMind, London, UK; Imperial College London, London, UK.
Imperial College London, London, UK.
Trends Cogn Sci. 2020 Nov;24(11):862-872. doi: 10.1016/j.tics.2020.09.002. Epub 2020 Oct 8.
The problem of common sense remains a major obstacle to progress in artificial intelligence. Here, we argue that common sense in humans is founded on a set of basic capacities that are possessed by many other animals, capacities pertaining to the understanding of objects, space, and causality. The field of animal cognition has developed numerous experimental protocols for studying these capacities and, thanks to progress in deep reinforcement learning (RL), it is now possible to apply these methods directly to evaluate RL agents in 3D environments. Besides evaluation, the animal cognition literature offers a rich source of behavioural data, which can serve as inspiration for RL tasks and curricula.
常识问题仍是人工智能发展的主要障碍。在这里，我们认为人类的常识建立在一系列基本能力之上，而许多其他动物也拥有这些能力，包括对物体、空间和因果关系的理解。动物认知领域已经开发出许多用于研究这些能力的实验方案，并且由于深度强化学习（RL）的进展，现在可以直接应用这些方法来评估 3D 环境中的 RL 代理。除了评估之外，动物认知文献还提供了丰富的行为数据来源，可以为 RL 任务和课程提供灵感。
