Zhun Deng

Hi! I am a postdoctoral researcher with Toniann Pitassi and Richard Zemel at Columbia University, and also part of Simons Collaboration on the Theory of Algorithmic Fairness. Previously, I completed my Ph.D. in the Theory of Computation group at Harvard University, advised by Cynthia Dwork. I am also fortunate to work with David Parkes, Weijie Su, and James Zou on various projects. 

My research investigates reliable and responsible machine learning for decision making and social problems, especially from a rigorous theoretical perspective. In particular, I develop formal frameworks to address algorithmic and societal challenges in modern data science by leveraging techniques in distribution-free uncertainty quantification, (multi-)calibration, and reinforcement learning. 

News:

• 09/2023 - two papers accepted by NeurIPS 2023: distribution-free societal dispersion control (spotlight, top 3% among submissions) and uncertainty quantification in physics-informed nets. 

• 05/2023 - our paper about generalization theory for information bottleneck is accepted by ICML 2023.

• 01/2023 - two papers accepted by AISTATS 2023: reinforcement learning with stepwise fairness constraints and understanding multimodal contrastive learning and incoportate paired data.

• 01/2023 - two papers accepted by ICLR 2023: FIFA: making fairness more generalizable on imbalanced data and distribution-free quantile risk control.

• 11/2022 - our paper about a generalized multi-calibration method is accepted by ITCS 2023.

• 05/2022 - two papers accepted by ICML 2022: how does Mixup help calibration and algorithmic robustness implies better generalization (long presentation, top 2% among submissions).

• 04/2022 - our journal paper about the dynamics of learning nonlinear representations is published in Neural Computation.

• 03/2022 - I will join Columbia University as a postdoc in Fall, 2022.

• 03/2022 - two talks about scaffolding sets at Simons Institute for the Theory of Computing, UC Berkeley, and adversarial transfer learning at Tsinghua University. 

• 01/2022 - our paper about optimization landscapes of neural collapse is accepted by ICLR 2022.

• 12/2021 - defended my Ph.D. dissertation thesis.