Hi, I am Alyssa Li Dayan (pronounced “A-lee-sa Lee Day-an”); I also go by Li.
I am a Computer Science Ph.D. student at Berkeley Artificial Intelligence Research where I am advised by Stuart Russell. Before that, I was an Machine Learning (ML) research scientist at Uber ATG. I completed my BS at MIT where I did research with the Flavell, Cocosci and MIT-IBM Watson AI labs.
My research is supported by the Hertz Fellowship.
Interests
- Hierarchical reinforcement learning and skill discovery
- Continual and lifelong Learning
- Meta learning
- Exploration and intrinsic motivation
Education
PhD in Computer Science, 2025 (expected)
UC Berkeley
BS in Math with Computer Science, 2018
Massachusetts Institute of Technology
Recent Publications
(2022).
Hierarchical Abstraction for Combinatorial Generalization in Object Rearrangement.
ICLR 2023.
(2021).
A Neural Circuit for Flexible Control of Persistent Behavioral States.
eLife 10.
(2021).
Multixnet: Multiclass Multistage Multimodal Motion Prediction.
In 2021 IEEE Intelligent Vehicles Symposium (IV).
(2019).
Analyzing Sketches Around the World with sketch-rnn.
Analytics Vidhya, results selected for display at the AI Art Gallery at NeurIPS 2020.
Contact
- dayan at berkeley dot edu