Computational Modeling of Learning in Complex Environments
Anne Collins, Professor
Psychology
Closed. This professor is continuing with Spring 2024 apprentices on this project; no new apprentices needed for Fall 2024.
Navigating complex natural environments presents a robust challenge to even the most advanced artificial systems, yet humans can often handle these same challenges effortlessly. What algorithms underly these human capabilities to learn and generalize so efficiently? My research investigates the learning algorithms implemented by the human brain to efficiently act in complex environments where the features of interest are unknown and the space of possible actions is large. Through these investigations we hope to both gain greater insight into human behavior and help move ideas from psychology lab experiments towards more realistic problems faced in the real world.
Ongoing projects focus on generalization and exploration in environments with complex high dimensional perceptual or action spaces, using methods from cognitive science, computer science, deep learning, and statistics. Apprentices will learn to implement behavioral experiments, analyze human behavioral data, and perform relevant computational modeling. Desired skills include programming (Python, MATLAB, and HTML/CSS/Javascript), data analysis and visualization, and basic knowledge about probability and statistics. Additional experience working within deep learning frameworks (tensorflow, pytorch) appreciated but not required.
Role: Students will be expected to familiarize themselves with the project and read suggested literature so that they can learn more about the hypotheses and goals of the research. They will also be expected to participate in the more general lab experience, with regular group or lab meetings with other lab members and the principal investigator Dr. Anne Collins.
Qualifications: Applicants should have strong interests in cognitive science, and some background in computer science and statistics. Experience with data analysis and visualization is a plus. We also have a preference for students who expect to volunteer in the lab for more than a single academic year. Students must be motivated, organized, and communicative.
Day-to-day supervisor for this project: Daniel Ehrlich, Post-Doc
Hours: 12 or more hours
Related website: https://ccn.berkeley.edu/
Social Sciences Education, Cognition & Psychology