Shelvia Wongso

Researcher in Information Theory for Machine Learning.

prof_pic.jpg

I am a self-driven researcher with a strong passion for exploring innovative ideas, solving complex problems, and always ready to take on new challenges. I recently earned my PhD in Artificial Intelligence (AI) from the National University of Singapore (NUS), where I was supervised by Professor Mehul Motani. In my thesis, I investigated how information-theoretic measures can be used to analyze deep neural networks. My research areas in deep learning include generalization, explainability, fairness, robustness, uncertainty quantification and privacy, all of which are important to build safe and trustworthy AI models.

Other interests: particle physics, neuroscience, psychology
Hobbies: exercising, reading books, gaming, and journalling

Contact: shelvia@nus.edu.sg

News

Nov 22, 2024 I passed my thesis defense yay! Hoping to publish my thesis once I finish refining it. The title of my thesis is “Generalization and Trustworthiness in Deep Learning Through the Lens of Information-Theoretic Measures”.
May 7, 2024 My abstract is accepted at Recent Results Poster Session of the ISIT 2024.
Apr 14, 2024 My paper is accepted at ISIT 2024 Workshop on Information-Theoretic Methods for Trustworthy Machine Learning.

Latest Posts

Selected Publications

  1. Pointwise Sliced Mutual Information for Neural Network Explainability
    Shelvia Wongso, Rohan Ghosh, and Mehul Motani
    In IEEE International Symposium on Information Theory, (ISIT), 2023
  2. Using Sliced Mutual Information to Study Memorization and Generalization in Deep Neural Networks
    Shelvia Wongso, Rohan Ghosh, and Mehul Motani
    In International Conference on Artificial Intelligence and Statistics (AISTATS), 2023
    Oral Presentation at AISTATS (Top 1.9% of Submitted Papers).
  3. Understanding Deep Neural Networks Using Sliced Mutual Information
    Shelvia Wongso, Rohan Ghosh, and Mehul Motani
    In IEEE International Symposium on Information Theory, (ISIT), 2022