Fast Concept Mapping: The Emergence of Human Abilities in Artificial Neural Networks when Learning Embodied and Self-Supervised
This work addresses the resource-intensive nature of fully supervised learning for AI agents by proposing a more human-like, self-supervised approach to concept acquisition, showing strong performance with minimal labels.
This paper explores a self-supervised learning setup where an artificial agent learns through embodied exploration in a simulated world. The learned representations are then used with "fast concept mapping" to associate semantic concepts, enabling object identification with as little as one labeled example.
Most artificial neural networks used for object detection and recognition are trained in a fully supervised setup. This is not only very resource consuming as it requires large data sets of labeled examples but also very different from how humans learn. We introduce a setup in which an artificial agent first learns in a simulated world through self-supervised exploration. Following this, the representations learned through interaction with the world can be used to associate semantic concepts such as different types of doors. To do this, we use a method we call fast concept mapping which uses correlated firing patterns of neurons to define and detect semantic concepts. This association works instantaneous with very few labeled examples, similar to what we observe in humans in a phenomenon called fast mapping. Strikingly, this method already identifies objects with as little as one labeled example which highlights the quality of the encoding learned self-supervised through embodiment using curiosity-driven exploration. It therefor presents a feasible strategy for learning concepts without much supervision and shows that through pure interaction with the world meaningful representations of an environment can be learned.