David W. Zhang

David W. Zhang

AI/ML Researcher

Qualcomm AI

Biography

I’m a machine learning researcher at Qualcomm AI Research in Amsterdam. Previously, I did my PhD in machine learning at the University of Amsterdam under the supervision of Cees Snoek and Gertjan Burghouts. I was also part of the ELLIS PhD program in cooperation with TNO. My research interests include structured prediction and generative models for structured data such as sets, graphs, or images.

Interests
  • Machine Learning
  • Computer Vision
Education
  • PhD in Machine Learning, expected 2023

    University of Amsterdam

  • MSc in Computer Science, 2017

    Technical University of Munich

  • BSc in Computer Science, 2015

    Technical University of Munich

Publications

Neural Networks Are Graphs! Graph Neural Networks for Equivariant Processing of Neural Networks

2nd Annual Topology, Algebra, and Geometry in Machine Learning Workshop at ICML (TAG-ML), 2023.
Unlocking Slot Attention by Changing Optimal Transport Costs

International Conference on Machine Learning (ICML), 2023.
Self-Guided Diffusion Models

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023.
Robust Scheduling with GFlowNets

International Conference on Learning Representations (ICLR), 2023.
Multiset-Equivariant Set Prediction with Approximate Implicit Differentiation

International Conference on Learning Representations (ICLR), 2022.
Pruning Edges and Gradients to Learn Hypergraphs from Larger Sets

Learning on Graphs (LoG), 2022.
Set Prediction without Imposing Structure as Conditional Density Estimation

International Conference on Learning Representations (ICLR), 2021.

Industry experience

 
 
 
 
 
Research Intern
Qualcomm AI Research
Jun 2022 – Sep 2022 Amsterdam, Netherlands
Research machine learning methods for computation graph scheduling.
 
 
 
 
 
Consultant
KPMG
Jan 2018 – Mar 2019 Munich, Germany