Neuro-Symbolic Concepts
This addresses the challenge of creating adaptable AI agents for various domains, though it appears incremental as it builds on existing neuro-symbolic approaches.
The paper tackles the problem of building agents that can learn continually and reason flexibly by introducing a concept-centric paradigm using neuro-symbolic concepts, resulting in advantages like data efficiency, compositional generalization, and zero-shot transfer across domains such as 2D images, videos, 3D scenes, and robotic manipulation.
This article presents a concept-centric paradigm for building agents that can learn continually and reason flexibly. The concept-centric agent utilizes a vocabulary of neuro-symbolic concepts. These concepts, such as object, relation, and action concepts, are grounded on sensory inputs and actuation outputs. They are also compositional, allowing for the creation of novel concepts through their structural combination. To facilitate learning and reasoning, the concepts are typed and represented using a combination of symbolic programs and neural network representations. Leveraging such neuro-symbolic concepts, the agent can efficiently learn and recombine them to solve various tasks across different domains, ranging from 2D images, videos, 3D scenes, and robotic manipulation tasks. This concept-centric framework offers several advantages, including data efficiency, compositional generalization, continual learning, and zero-shot transfer.