Jordan Lei
Neuroscience | Machine Learning | Art
About
Hi, I'm Jordan! I'm a PhD Candidate in the Wei Ji Ma Lab at New York University, where I study planning and decision making in the brain.
Here are some questions I'm thinking about: how do people respond to uncertainty when planning? Do people think differently than machines when playing chess? How does the brain think about the future?
I'm passionate about the intersection of neuroscience and deep learning. Outside work, you'll catch me sketching in the city (probably at the Met), running in Manhattan, or snuggling up with a good book.
Research
Can Monkeys Play Board Games?
Planning is hard. How does the brain do it? We collaborated with Lee Lab at Johns Hopkins University to understand how monkeys plan when playing a complex planning game, Four-in-a-Row, against a computer opponent. We are building a model that combines gameplay, eye movement, and neural electrophysiological data to understand the neural mechanisms of planning.
View PosterHow Does Uncertainty Affect Planning Effort?
The future is often uncertain, which influences planning behavior. We designed an online task with three different types of uncertainty and modeled participant planning strategy. We found that people decreased their planning depth in response to increased uncertainty - intuitively, if the future is very uncertain, why bother planning deeply?
View PaperView PosterHow Do People Think Ahead in Chess?
For hundreds of years, chess has been associated with intelligence, strategy, and complex reasoning. Recent advances in artificial intelligence and the popularity of online chess make it an exciting time to study chess from a cognitive science perspective. In this project, we are building a cognitive model of how people think ahead in chess.
Experience
View ResumePhD Candidate (Researcher) - Ma Lab, New York University
Aug 2021 - Present • New York, NY
- My research focus is understanding the neural and cognitive mechanisms of complex planning in humans. I use reinforcement learning and deep learning models to understand how people and animals think ahead. For more information see Research.
- Awards: 2023 Training Program in Computational Neuroscience Grant, 2021 IVADO PhD Excellence Scholarship (declined), 2021 Henry M. MacCracken Fellowship
Researcher - Kording Lab & Gold Lab, University of Pennsylvania
May 2020 - May 2021 • Philadelpha, PA
- Created a deep learning model of visual attention. Incorporated convolutions, recurrence, encoder-decoder architectures, and custom loss functions to build a model that replicates key features of biological attention, including inhibition of return and magnitude shifts in tuning curves. Work submitted as master's thesis.
- Compared biologically plausible and artificial learning algorithms. Analyzed common failure modes of biologically plausible Hebbian learning agents and backpropagation, such as catastrophic forgetting.
- Awards: Lila R. Gleitman MINDCORE Summer Fellowship
Finance Intern - Sales & Operations Planning, Unilever
May 2019 - Aug 2019 • Englewood Cliffs, NJ
- Worked with Sales and Operations Planning to speed up the cash flow reporting process. Created a full-stack web application in Python to automate reporting of statement of cash flows for Sales and Operations Planning - reduced cash flow reporting time by over 80% and made the process interoperable with Microsoft Excel.
Finance & Data Science Intern, Tovala
May 2018 - Aug 2018 • Chicago, IL
- Tovala is a startup in the smart devices and meal-delivery space. Estimated customer acquisition costs and analyzed the efficacy of their online advertising and created a predictive model of packaging costs.
Education
New York University
PhD Candidate in Neuroscience
In Progress
Thesis | Neural and Cognitive Mechanisms of Planning
Advisor | Wei Ji Ma
Teaching | NEURL-GA.2201 Mathematical Tools for Neuroscience
University of Pennsylvania
MSE in Computer Science
Class of 2021, Summa Cum Laude
Thesis | Object-based Attention Through Internal Gating
Advisor | Konrad Kording
Teaching | CIS 522 Deep Learning (Lead TA)
University of Pennsylvania
BS in Economics, Operations, Info, & Decisions
BS in Engineering, Computer Science
Class of 2020, Summa Cum Laude
Jerome Fisher Program of Management & Technology
Teaching | CIS 519 Machine Learning