The ability to discover abstract physical concepts and understand how they work in the world through observing lies at the core of human intelligence. The acquisition of this ability is based on compositionally perceiving the environment in terms of objects and relations in an unsupervised manner. Recent approaches learn object-centric representations and capture visually observable concepts of objects, e.g., shape, size, and location. In this paper, we take a step forward and try to discover and represent intrinsic physical concepts such as mass and charge. We introduce the \uppercase{phy}sical \uppercase{c}oncepts \uppercase{i}nference \uppercase{ne}twork (PHYCINE), a system that infers physical concepts in different abstract levels without supervision. The key insights underlining PHYCINE are two-fold, commonsense knowledge emerges with prediction, and physical concepts of different abstract levels should be reasoned in a bottom-up fashion. Empirical evaluation demonstrates that variables inferred by our system work in accordance with the properties of the corresponding physical concepts. We also show that object representations containing the discovered physical concepts variables could help achieve better performance in causal reasoning tasks, i.e., ComPhy.

翻译：通过观察发现抽象物理概念并理解它们在世界中如何运作的能力，是人类智能的核心。这种能力的获得基于以对象和关系为单元、以无监督方式对环境进行组合性感知。近期方法学习面向对象的表征，并捕捉对象在视觉上可观察的概念（例如形状、大小和位置）。在本文中，我们向前迈进一步，尝试发现并表征内在物理概念，例如质量和电荷。我们引入物理概念推理网络（PHYCINE），该系统无需监督即可在不同抽象层次上推断物理概念。支撑PHYCINE的关键洞见有两点：常识知识伴随预测而涌现，且不同抽象层次的物理概念应以自底向上的方式进行推理。实验评估表明，我们系统推断出的变量与相应物理概念的性质保持一致。我们还发现，包含所发现的物理概念变量的对象表征有助于在因果推理任务（即ComPhy）中取得更优性能。