[Please refer to my CV/github for new updates. I have been lazy updating this site...]
I am currently pursuing PhD in CS at University of Southern California, working on Domain shift/Robustness/OOD/fairness topics in Machine Learning, under the supervision of Prof. Xuezhe Ma. Moreover, I am fortunate to work with Dr. Pin-Yu Chen on Robust Machine Learning.
I encourage aspiring undergraduate and Master's studetns to contact me for research opportunities.
My current research interests lie in but not limited to topics on Domain shift/Robustness/OOD/fairness. Ultimately want to learn causal representations! I am also interested in commercial applications of AI such as content generation, drug discovery, agricultural resource allocation, etc.
Previously, I had the pleasure to work with Prof. Muhammad Naveed on Adversarial ML, Prof. Xiang Ren and Prof. Shrikanth Narayanan on determining Screenplay Quality with textual properties. And with Prof. Stefan Scherer, we designed an conversation quality assessment system using small segments of text.
More to come..
On Human Visual Contrast Sensitivity and Machine Vision Robustness: A Comparative Study (submitted to CVPR 2023) (preprint)
Better May Not Be Fairer: Can Data Augmentation Mitigate Subgroup Degradation? (submitted to CVPR 2023) (preprint)
Robust Representation Learning for Group Shifts and Adversarial Examples. (NeruIPS ML Safety 2022) (preprint)
Screenplay Quality Assessment: Can We Predict Who Gets Nominated? In Proceedings of the ACL Workshop on Narrative Understanding, Storylines, and Events (NUSE 2020), July 2020
Automated Empathy Detection for Oncology Encounters. In Proceedings of the IEEE International Conference on Healthcare Informatics (ICHI2020), December 2020.
A system for the 2019 Sentiment, Emotion and Cognitive State Task of DARPA's LORELEI project. In Proceedings of the International Conference on Affective Computing & Intelligent Interaction (ACII 2019), September 2019. (Oral Presentation)
Silver Medal (top 4% out of 5,332 teams) in Kaggle Porto Seguro’s Safe Driver Prediction competition, Nov. 2017
Attended The Data Open Datathon, Pasadena (80 out of over 600 contestants) sponsored by Citadel, Oct. 2017
Viterbi Art Competition, Ming Hsieh Institute -- Attending Prize, Sep. 2017