Home | Joel Ye

(Incoming) PhD student at CMU

Say hello: joelye9 at gmail dot com

Bio

I study machine learning as a Master's student at Georgia Tech. My research aims to relate computation in the brain and in AI systems, with a eye towards applications in brain-computer interfaces. I'm currently working with Dhruv Batra, Erik Wijmans, and Abhishek Das on embodied navigation. I also work with Chethan Pandarinath on modeling neural data.

I grew up in NYC, where I graduated from Stuyvesant High School, and I did my undergrad in Computer Science at Georgia Tech. During my undergrad, I was Director of Technology for Georgia Tech's hackathon org, HackGT, where we worked to host events to promote CS education.

Hobbies I make time for include reading (and am always open to recommendations) and studying Japanese and Chinese.

News

[5/4/21] Course project writeups for learning from demonstration and computation and the brain posted.
[4/8/21] I will be beginning my PhD in Neural Computation at CMU in the Fall!
[4/8/21] I will be interning with Dr. Robinson Piramuthu as an Applied Scientist intern at Amazon for Summer 21.
[3/31/21] Awarded the Donald V. Jackson Fellowship (for 1st year MS students).
[12/4/20] Network Science course project writeup posted here. I suggest we may want to study dynamical robustness in deep networks.
[11/25/20] Deep Learning for Text course project writeup posted here. We looked at how BERT's representations change during fine-tuning.

Projects

Auxiliary Tasks and Exploration Enable ObjectNav

Joel Ye, Dhruv Batra, Abhishek Das, Erik Wijmans

arXiv 2021

Auxiliary Tasks and Exploration Enable ObjectNav

Neural Data Transformers

Joel Ye, Chethan Pandarinath

In review (preprint)

Auxiliary Tasks Speed Up Learning PointGoal Navigation

Joel Ye, Dhruv Batra, Erik Wijmans*, Abhishek Das*

CoRL, 2020

Auxiliary Tasks Speed Up Learning PointGoal Navigation