Khushbu Pahwa

|| "कर्मण्येवाधिकारस्ते मा फलेषु कदाचन। मा कर्मफलहेतुर्भूर्मा ते संगोऽस्त्वकर्मणि॥ || - "To perform one's duties diligently and with dedication without attachment to the outcomes"

prof_pic2.png

Duncan Hall, 3014r

6100 Main Street

Houston, TX 77005

I’m Khushbu Pahwa, a recent Masters in Electrical & Computer Engineering Graduate from University of California, Los Angeles.

My research interests include efficient and trustworthy machine learning for foundation models, with special interest towards healthcare applications. I am broadly interested in the domain of Multi-Modal Learning, Robust Deep Learning, and Trustworthy AI

During my time at UCLA, I was fortunate to engage in challenging research projects at UCLA with Dr. Abeer Alwan (Depression detection from speech while preserving speaker identity - [ Privacy Preserving ML ] ), Dr Dan Ruan (fast and learnable measurement-conditioned undersampled MR image reconstruction using Diffusion models [ Efficient Foundation Model Research using Diffusion Models]), Dr. Nader Sehatbaksh (comprehensively evaluating the adversarial robustness of tiny ML models - [ Neural Architecture Search Framework for adversarially robust Tiny-ML Models ] ); and with Dr. Cho-Jui Hsieh (developing computationally efficient gradient based white-box adversarial attack against text transformers). During the summer of 2022, I interned at Amazon as an Applied Scientist Intern in the AWS Machine Learning Solutions Lab.

Additionally, I have had an incredible opportunity to engage in research works with Prof. Pengtao Xie (UCSD) , Prof. Amitava Das (AI Institute at UoSC) , Prof. Sourav Medya (UIC) , and Dr. Manish Gupta (Microsoft R&D) .

Previously, I received my bachelor’s degree in Delhi Technological University, where I was awarded the Vice Chancellor Gold Medal for academic and research excellence.

Contact me at: khushbu16pahwa@g.ucla.edu .

News

Oct 7, 2023 1 research paper accepted at EMNLP 2023: FACTIFY3M: A benchmark for multimodal fact verification with explainability through 5W Question-Answering :sparkles: :smile:
Oct 4, 2023 Research Paper : “GnnX-Bench: Unravelling the Utility of Perturbation-based GNN Explainers through In-depth Benchmarking” is live on ArXiv
Aug 21, 2023 Awarded the Ken Kennedy Institute Computational Science & Engineering Recruiting Fellowship Announcement:tada:
Jul 8, 2023 Scaling Distributed Multi-task Reinforcement Learning with Experience Sharing accepted for poster presentation at KDD 2023 Federated Learning Workshop :sparkles: :smile:
Jun 20, 2023 Graduated with MS in ECE from UCLA. Self supported education with TA and GSR for each of the 6 quarters :tada:
Feb 15, 2023 Research Paper “Neural Architecture of Speech” accepted for Oral Presentation at ICASSP 2023 :sparkles: :smile:

Selected Publications

Please refer to the publications