Mijung’s primary focus is on building privacy-preserving machine learning algorithms using a sophisticated mathematical notion called differential privacy. Her other research interests include: compressing neural network models using Bayesian methods, and; understanding the relationships between differential privacy and other emerging notions in machine learning such as fairness, interpretability, and causality.
- Harder, F., Adamczewski, Park, M. DP-MERF: Differentially Private Mean Embeddings with Random Features for Practical Privacy-preserving Data Generation, Artificial Intelligence and Statistics 2021, PMLR 130:1819‑1827
- Adamczewski K., Park, M. Dirichlet Pruning for Convolutional Neural Networks, Artificial Intelligence and Statistics 2021, PMLR 130:3637‑3645
- Harder, F., Bauer, M., Park, M. Interpretable and Differentially Private Predictions, AAAI Conference on Artificial Intelligence 2020, 34(04), 4083‑4090.
- Park M., Foulds, J., Chaudhuri, K., Welling, M. Variational Bayes in Private Settings (VIPS), Journal of Artificial Intelligence Research 2020.
- Park, M., Foulds, J., Chaudhuri, K., Welling, M. DP-EM: Differentially Private Expectation Maximization, Artificial Intelligence and Statistics (AISTATS) 2017, PMLR 54:896‑904.
CIFAR is a registered charitable organization supported by the governments of Canada, Alberta and Quebec, as well as foundations, individuals, corporations and Canadian and international partner organizations.