非表示:
キーワード:
-
要旨:
Most cognitive models are domain-specific, meaning that their scope is restricted to a single type of problem. The human mind, on the other hand, does not work like this -- it is a unified system whose processes are deeply intertwined. My research closes this gap by taking a domain-general perspective on human cognition. In the first part of this talk, I will outline how the framework of meta-learning can be used to construct cognitive models across various domains. In the second part, I will summarize my recent work on the intersection between large language models and cognitive science. In the third and final part, I will present a blueprint for how ideas from meta-learning and large language models can be combined to create foundation models of human cognition -- models that cannot only simulate, predict, and explain human behavior in a single domain but that offer a unified take on our mind.
