What makes humans so uniquely intelligent?
How do people make the best use of limited cognitive resources?
What are the unique algorithms we use to learn from other people?
Led by Charley Wu, the Human and Machine Cognition (HMC) Lab works at the intersection of Human Cognitive Science and Machine Learning research. We seek to use insights from human cognition to improve machine learning methods, while also using advances in machine learning as tools for understanding human intelligence. In practice, this often involves using experiments to study an interesting feature of human cognition (e.g., rapid learning in novel environments) and then using computational models to understand the cognitive principles behind the phenomenon.
Experiments are typically performed online in the form of an interactive game, in the lab using computers or virtual reality equipment, or occur naturally through the analysis of real-world data. Our computational models are often inspired by reinforcement learning or machine learning methods, and are validated by predicting human behavior, neural activity, or through evolutionary simulations.
In collaboration with Peter Dayan from the Max Planck Institute for Biological Cybernetics, we are currently seeking a Postdoc to study Representation Learning