Rares-Darius Buhai

I am a postdoc in the School of Computer and Communication Sciences at EPFL, hosted by Prof. Ola Svensson. I am interested in theoretical aspects of computer science and machine learning.

Email: rares (-) darius (.) buhai (@) epfl (.) ch.

Bio

• 2025-present: postdoc, EPFL, hosted by Prof. Ola Svensson
• 2020-2025: PhD, ETH Zurich, advised by Prof. David Steurer
• 2015-2020: BSc & MEng, MIT, advised by Prof. Guy Bresler

Publications

• Finding Colorings in One-Sided Expanders
  RB, Yiding Hua, David Steurer, Andor Vári-Kakas
  FOCS 2025
• The Quasi-Polynomial Low-Degree Conjecture is False
  RB, Jun-Ting Hsieh, Aayush Jain, Pravesh K. Kothari
  FOCS 2025
• Lasso and Partially-Rotated Designs
  RB
  preprint
• Dimension Reduction via Sum-of-Squares and Improved Clustering Algorithms for Non-Spherical Mixtures
  Prashanti Anderson, Mitali Bafna, RB, Pravesh K. Kothari, David Steurer
  preprint
• Robust Mixture Learning when Outliers Overwhelm Small Groups
  Daniil Dmitriev, RB, Stefan Tiegel, Alexander Wolters, Gleb Novikov, Amartya Sanyal, David Steurer, Fanny Yang
  NeurIPS 2024
• Semirandom Planted Clique and the Restricted Isometry Property
  Jarosław Błasiok, RB, Pravesh K. Kothari, David Steurer
  FOCS 2024
• Computational-Statistical Gaps for Improper Learning in Sparse Linear Regression
  RB, Jingqiu Ding, Stefan Tiegel
  COLT 2024
• Beyond Parallel Pancakes: Quasi-Polynomial Time Guarantees for Non-Spherical Gaussian Mixtures
  RB, David Steurer
  COLT 2023
• Algorithms approaching the threshold for semi-random planted clique
  RB, Pravesh K. Kothari, David Steurer
  STOC 2023
• Learning Restricted Boltzmann Machines with Sparse Latent Variables
  Guy Bresler, RB
  NeurIPS 2020
• Empirical Study of the Benefits of Overparameterization in Learning Latent Variable Models
  RB, Yoni Halpern, Yoon Kim, Andrej Risteski, David Sontag
  ICML 2020