Jin Hwa Lee
Hi, I’m Jin (she/her)!
I am a PhD student at University College London supervised by Andrew Saxe.
At the moment, I am mainly intrigued by compositionality in learning and inference. While compositionality is broad and very general capability that humans have in language acquisition, skill learning and generalization problems, current artificial intelligent systems do not exhibit the same level of compositional ability.
How do humans discover useful motifs or concepts and flexibly reuse them to solve complex tasks?
What is the neural mechanism that supports compositionality?
Can we build an artificial intelligence system which exploits compositionality in learning and inference?
To answer these questions, I aim to study both biological brains and artificial neural networks.
I use artificial neural networks to model interesting cognitive/behavioural observations in humans and animals to bring quantitative understanding of the neural mechanisms behind.
Before starting my PhD journey, I completed my master’s degree in TU Munich. I was supervised by Mackenzie Mathis and developed CEBRA for my master’s thesis project.
See Research for more on previous/ongoing projects.
news
Jul 20, 2024 | I’m giving a tutorial talk in Conference on Lifelong Learning Agents! in August, Pisa. I’ll be talking about theoretical approaches used to study continual learning. |
---|---|
Jul 1, 2024 | … Also a Mech. Interp workshop paper in ICML 2024: Modularity in Biologically Inspired Representations Depends on Task Variable Range Independence |
Jul 1, 2024 | … Also a Mech. Interp workshop paper in ICML 2024: Modularity in Biologically Inspired Representations Depends on Task Variable Range Independence |
Jul 1, 2024 | My first PhD project Why do Animals Need Shaping? A Theory of Task Composition and Curriculum Learning is accepted to ICML 2024! |
Sep 30, 2023 | I joined Saxe Lab and started a PhD journey 🚀🧠 :)) |