About Me

I completed my PhD at the Hebrew University of Jerusalem in October 2023 as part of the ELSC International Graduate Program, under the supervision of Prof. Daphna Weinshall. My PhD thesis focused on studying the learning dynamics of deep networks from a data-centric perspective. Currently, I am a post-doctoral researcher at KU Leuven in the ESAT-PSI lab, working with Prof. Tinne Tuytelaars. My current research extends into continual learning, exploring both the learning and the forgetting dynamics of deep networks.

Publications

• Forgetting Order of Continual Learning: Examples That are Learned First are Forgotten Last
  Guy Hacohen and Tinne Tuytelaars
  arXiv 2024
  [arXiv]
• Pruning the Unlabeled Data to Improve Semi-Supervised Learning
  Guy Hacohen and Daphna Weinshall
  arXiv 2023
  [arXiv]
• Semi-Supervised Learning in the Few-Shot Zero-Shot Scenario
  Noam Fluss, Guy Hacohen and Daphna Weinshall
  arXiv 2023
  [arXiv]
• How to Select Which Active Learning Strategy is Best Suited for Your Specific Problem and Budget
  Guy Hacohen and Daphna Weinshall
  NeurIPS 2023
  [arXiv]
• Active Learning Through a Covering Lens
  Ofer Yehuda, Avihu Dekel, Guy Hacohen and Daphna Weinshall
  NeurIPS 2022
  [arXiv]
• Active Learning on a Budget: Opposite Strategies Suit High and Low Budgets
  Guy Hacohen, Avihu Dekel and Daphna Weinshall
  ICML 2022
  [arXiv] [Blog post] [GitHub] [Spotlight]
• The Grammar-Learning Trajectories of Neural Language Models
  Leshem Choshen, Guy Hacohen, Daphna Weinshall, Omri Abend
  ACL, 2022.
  [arXiv]
• Principal Components Bias in Over-parameterized Linear Models, and its Manifestation in Deep Neural Networks
  Guy Hacohen and Daphna Weinshall
  JMLR 2022
  [arXiv] [60 minutes lecture]
• Let's Agree to Agree: Neural Networks Share Classification Order on Real Datasets
  Guy Hacohen, Leshem Choshen and Daphna Weinshall
  ICML, 2020.
  [arXiv] [Oral Presentation]
• On The Power of Curriculum Learning in Training Deep Networks
  Guy Hacohen and Daphna Weinshall
  ICML, 2019.
  [arXiv] [GitHub] [Spotlight]