Research

. Large Language Models Can Be Strong Differentially Private Learners. Oral presentation at NeurIPS Privacy in Machine Learning Workshop (PriML’21), 2021.

PDF Code

. On the Opportunities and Risks of Foundation Models. 2021.

PDF

. Efficient and Accurate Gradients for Neural SDEs. Advances in Neural Information Processing Systems (NeurIPS), 2021.

PDF

. Neural SDEs as Infinite-Dimensional GANs. International Conference on Machine Learning (ICML), 2021.

PDF

. When Does Preconditioning Help or Hurt Generalization?. Best Student Paper at the 12th OPT Workshop on Optimization for Machine Learning. International Conference on Learning Representations (ICLR), 2021.

Preprint

. Scalable Gradients for Stochastic Differential Equations. Spotlight presentation at the 2nd Symposium on AABI. International Conference on Artificial Intelligence and Statistics (AISTATS), 2020.

PDF Code Video Slides

. Stochastic Runge-Kutta Accelerates Langevin Monte Carlo and Beyond. Spotlight presentation at Advances in Neural Information Processing Systems (NeurIPS), 2019.

PDF slides

. The idemetric property: when most distances are (almost) the same. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2019.

PDF

. Isolating Sources of Disentanglement in Variational Autoencoders. Oral presentation at Advances in Neural Information Processing Systems (NeurIPS), 2018.

PDF video

. Inference Suboptimality in Variational Autoencoders. International Conference on Machine Learning (ICML), 2018.

PDF