Chris Yuhao Liu

yliu298 [at] ucsc [dot] edu

I am an incoming Master's student at the University of California, Santa Cruz, where I also did my undergrad in computer science. I also work as a researcher at Professor Jeffrey Flanigan’s JLab.

I am currently working on algorithms that can improve the sample complexity rate of machine learning algorithms. I am also interested in learning with less data, out-of-distribution generalization, interpretability of neural networks, and neuroscience.

Blog  /  CV  /  Email  /  Github  /  LinkedIn

• [2020-12] Here is my blog page.
• [2020-11] I collected a list of useful resources for computer science graduate school application.
• [2020-09] I am excited to tutor the course CSE142 Machine Learning at UC Santa Cruz in the Fall 2020 quarter.
• [2020-06] I joined Professor Jeffrey Flanigan's JLab.

These included publications and preprints.

Faster Sample Complexity Rates With Ensemble Filtering
Chris Yuhao Liu, Jeffrey Flanigan
2021
In submission

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.

These include coursework and side projects.

TAPT: Text Augmentation Using Pre-Trained Transformers With Reinforcement Learning
UC Santa Cruz
2020-07

I fine-tuned a distilled RoBERTa model as a text classifier on the IMDb Large Movie Review Dataset and a GPT-2 (345M) as a text generator using the proximal policy optimization (PPO) framework.

Fine-Tuning GPT-2 to Generate Research Paper Abstracts
UC Santa Cruz
2020-06

I fine-tuned a pre-trained GPT-2 (774M) model using all research paper titles and abstracts under cs.AI, cs.LG, cs.CL, and cs.CV on arXiv and built a machine learning paper abstract generator. The model is hosted by Hugging Face. This project was the winner of the Image/Text Generation Competition for the course CSE142 Machine Learning in Spring 2020.

Sentiment Analysis With Transformers
UC Santa Cruz
2020-06

I fine-tuned a RoBERTa (355M) model using the IMDb dataset and achieved 96.516% accuracy. This project was the winner of the Sentiment Analysis Competition for the course CSE142 Machine Learning in Spring 2020.

• Volunteer, Ninth International Conference on Learning Representations (ICLR 2021)
• Tutor, CSE142 Machine Learning Fall 2020

This is a fork of Jon Barron's website.