Anne Collins, Professor

Closed (1) Applying deep learning methods to reinforcement learning models

Closed. This professor is continuing with Spring 2021 apprentices on this project; no new apprentices needed for Fall 2021.

Reinforcement learning (RL) lies at the intersection of multiple fields. Researchers have used RL in Psychology to explain human behavior, in Neuroscience to understand brain function, and in Computer Science to create state-of-the-art AI systems.

To fully leverage the power of RL in understanding human behavior, we often use traditional statistical methods such as maximum likelihood to fit RL models with human behavior. However, such methods fail for more complex RL models, which are essential for explaining more complex human cognition.

This project proposes to use modern deep learning methods to directly fit RL models instead.
The research assistant will have an opportunity to study and understand different families of RL models that are being extensively used in the frontier of RL research, and use deep learning techniques to fit RL models to generate crucial and novel insights to complex human cognition.

The research assistant will be expected to be familiar with the goals and the contents of the project, and will closely collaborate with the graduate student mentor. The mentor will provide guidance in reading some of the relevant literature, to put the specific project in a broader context.

The research assistant will be required to meet regularly with the mentor and to keep track of progress in the project.

The research assistant will have an opportunity to study and understand different families of RL models that are being extensively used in the frontier of RL research, and use deep learning techniques to fit RL models to generate crucial and novel insights to complex human cognition.

The research assistant will be expected to be familiar with the goals and the contents of the project, and will closely collaborate with the graduate student mentor. The mentor will provide guidance in reading some of the relevant literature, to put the specific project in a broader context.

The research assistant will be required to meet regularly with the mentor and to keep track of progress in the project.

Day-to-day supervisor for this project: Jimmy Xia, Graduate Student

Qualifications: To work on this project, experience and proficiency with commonly used modern deep learning techniques (any deep learning library is welcome) is required. Experience with Reinforcement Learning is preferred but not required.

Weekly Hours: to be negotiated

Off-Campus Research Site: Due to COVID, work on the project and meetings with the mentor will be done remotely.

Related website: https://www.ocf.berkeley.edu/~acollins/

Closed (2) EEG studies of inter-subject brain activity

Closed. This professor is continuing with Spring 2021 apprentices on this project; no new apprentices needed for Fall 2021.

Previous neurobiological research on social communication and exchange of information between two individuals revealed synchronization in neural activity across the brains of the communicating individuals. However, it is still unknown what is the exact meaning and purpose of this neural synchrony in human interaction.

This project aims to investigate that question using behavioral testing and electroencephalography (EEG). EEG is a non-invasive way of measuring large-scale brain activity through the scalp. Using EEG, we plan to further understand the purpose of the neural synchronization when transferring information from one person to another.


Research assistants will have an opportunity to participate in multiple stages of the research process and receive mentoring along the way. Research assistants will be required to meet regularly with their mentor and to keep track of their progress in the project.

Research assistants on this project will be responsible for data collection, neural data processing, and data analysis. They will be asked to prepare subjects for EEG data collection, assist in their participation in the behavioral task as well as pre-process and analyze the EEG data.

Furthermore, research assistants will be expected to familiarize themselves with the goals and contents of the project. They should be able to clearly explain the purpose and predictions of the experiment. The mentor will provide guidance in reading some of the relevant literature, to put the specific project in a broader context.


Day-to-day supervisor for this project: Liza Yartsev, Graduate Student

Qualifications: Applicants should be interested in cognitive neuroscience and willing to spend 9-11 hours a week on this project. We are looking for candidates with programming experience (MATLAB preferred, R and Python ok) and some basic statistical knowledge. Experience collecting human data is welcome but not necessary. The ideal candidate will be reliable, proactive, and highly motivated.

Weekly Hours: to be negotiated