Building Human-Like Artificial Agents: Advances from a Cognitive Computational Decision Science Perspective

Tune in for the live stream on YouTube or Twitter.

Abstract: One of the early goals of Artificial Intelligence (AI) was to create algorithms that exhibited behavior indistinguishable from human behavior (i.e., human-like behavior). Today, AI has diverged, often aiming to excel in tasks inspired by human capabilities and outperform humans, rather than replicating human actions. This talk explores the extent to which Cognitive Science has realized the original goal of AI by developing computational algorithms that mimic human decision making. Specifically, I will delve into the realm of dynamic decision making, approaching the question from the point of a comprehensive cognitive algorithm rooted in Instance-Based Learning Theory (IBLT), a cognitive theory of decisions from experience in dynamic environments. By utilizing the cognitive steps outlined within IBLT, I will discuss existing evidence supporting the human-likeness of decision-making mechanisms while identifying critical research gaps. The discussion seeks to shed light on the progress made in emulating human decision processes computationally, alluding to potential avenues for achieving greater fidelity in AI algorithms. Ultimately, I aim to advance the construction of machines that exhibit human-like behavior in dynamic environments so that we can support humans in making decisions in naturalistic tasks.