|
I am an MSc student in Computer Science and Engineering at the University of California, Santa Cruz. I also work as a researcher at Professor Jeffrey Flanigan’s JLab. My research interests broadly lie in memorization in deep neural networks, compositional generalization, the limite of deep learning, and computational neuroscience.My current research focuses on the scaling law between the amount of training data and the generalization performance of deep neural networks (aka the sample complexity rate/power-law). Previously, I obtained my B.S. in Computer Science and Engineering at UC Santa Cruz. |
|
|
|
These included publications and preprints. |
|
We present a dataset filtering approach that uses sets of classifiers, similar to ensembling, to estimate noisy (or non-realizable) examples and exclude them so a faster sample complexity rate is achievable in practice. |
|
We show that adversarially training (Fast Gradient Sign Method and Projected Gradient Descent) reduces the empirically sample complexity rate for MLP and a variety of CNN architectures on MNIST and CIFAR-10. |
|
These include coursework and side projects. |
|
A distilled RoBERTa model as a text classifier and a GPT-2 (345M) as a text generator trained using the proximal policy optimization (PPO) framework |
|
A GPT-2
(774M) trained using all research paper titles and
abstracts under cs.AI, cs.LG, cs.CL, and cs.CV on arXiv
|
|
A RoBERTa
(355M)
model using the IMDb
dataset
|
|
|
This is a fork of Jon Barron's website. |