Eran Treister

Eran Treister is an Assistant Professor at the Ben Gurion University of the Negev in the Department of Computer Science. He completed his postdoctoral fellowship at the University of British Columbia (2014-2016) and earned his PhD from the Technion in 2014 under Prof. Irad Yavneh.

His research spans computational science, numerical methods, and machine learning, with a focus on:

• Scalable algorithms for inverse problems
• Graph Neural Networks (GNNs) optimization
• Seismic and optical imaging via PDE solvers
• Low-precision deep learning acceleration
• Multilevel preconditioning techniques

Recent work explores:

• Graph neural networks for PDEs with adaptive meshes
• Deep learning approaches to Helmholtz equation modeling
• 3D shape reconstruction via parametric level sets

He serves on editorial boards:

• SIAM Journal on Scientific Computing (2024-)
• Copper Mountain Conference on Multigrid Methods (2025)
• International Conference on Machine Learning (ICML) as Area Chair (2025)

Current teaching:

• Optimization Methods for Data Science (Spring 2025)
• Deep Learning Mini-Project (Winter 2024/5)
• Advanced Numerical Optimization (Spring 2025)