David X. Wu

I am a second year Ph.D. student at UC Berkeley, where I’m extremely fortunate to be coadvised by Prasad Raghavendra and Anant Sahai. I’m broadly interested in problems at the intersection of theoretical computer science and statistics, including topics such as computational complexity for inference problems, Markov chains, machine learning theory, and optimization. I graduated with a B.Sc. from MIT with a double major in mathematics (Course 18) and computer science (Course 6-3). I’m grateful to be supported by an NSF GRFP fellowship.

selected publications

NeurIPS

Precise Asymptotic Generalization for Multiclass Classification with Overparameterized Linear Models

David X Wu, and Anant Sahai

Neural Information Processing Systems, 2023

arXiv Bib

@article{wu2023precise,
  title = {Precise Asymptotic Generalization for Multiclass Classification with Overparameterized Linear Models},
  author = {Wu, David X and Sahai, Anant},
  journal = {Neural Information Processing Systems},
  year = {2023},
}

ICML

On the Training Instability of Shuffling SGD with Batch Normalization

David X Wu, Chulhee Yun, and Suvrit Sra

International Conference on Machine Learning, 2023

arXiv Bib

@article{wu2023training,
  title = {On the Training Instability of Shuffling SGD with Batch Normalization},
  author = {Wu, David X and Yun, Chulhee and Sra, Suvrit},
  journal = {International Conference on Machine Learning},
  year = {2023},
}

SIOPT

Maximum A Posteriori Inference of Random Dot Product Graphs via Conic Programming

David X Wu, David Palmer, and Daryl R DeFord

SIAM Journal on Optimization, 2022

arXiv Bib

@article{wu2022maximum,
  title = {Maximum A Posteriori Inference of Random Dot Product Graphs via Conic Programming},
  author = {Wu, David X and Palmer, David and DeFord, Daryl R},
  journal = {SIAM Journal on Optimization},
  volume = {32},
  number = {4},
  pages = {2527--2551},
  year = {2022},
  publisher = {SIAM},
}