Yubei Chen

Postdoctoral Associate (Applying for a tenure-track position this year!)
Center for Data Science, New York University
Postdoctoral Associate, Advisor Professor Yann LeCun


I am a postdoctoral associate at NYU Center for Data Science (CDS) and Meta Fundamental AI Research (FAIR), working with Prof. Yann LeCun. Prior to joining CDS and FAIR, I completed my Ph.D. at Redwood Center for Theoretical Neuroscience and Berkeley AI Research (BAIR), UC Berkeley, advised by Prof. Bruno Olshausen. Undergraduate studies from Tsinghua University, Beijing. Co-founded Aizip Inc. that builds robust, efficient, and scalable AI-IoT solutions.


My research is at the intersection of computational neuroscience and deep unsupervised (self-supervised) learning, enhancing our understanding of the computational principles governing unsupervised representation learning in both brains and machines, and reshaping our insights into natural signal statistics.


news

Mar21, 2023
Jan21, 2023
Three papers are accepted at ICLR 2023 with one Oral and one Spotlight! :sparkle: 1) Oral: On the duality between contrastive and non-contrastive self-supervised learning arXiv. 2) Spotlight: Minimalistic unsupervised learning with the sparse manifold transform arXiv. 3) Simple Emergent Action Representations from Multi-task Policy Training arXiv.
Jan11, 2023
I gave talk at Center for Computational Neuroscience (CCN) Research Seminar, Flatiron Institute on the principles of unsupervised representation learning. :coffee:
Nov18, 2022
I gave a talk at University of Washington NeuroAI Seminar on the principles of unsupervised representation learning. :coffee:
Nov4, 2022
Simple Emergent Action Representations from Multi-Task Policy Training, is accepted to Deep RL Workshop@NeurIPS 2022! :sparkle: arXiv, Website
Oct21, 2022
Disentangling images with Lie group transformations and sparse coding, is accepted to NeurReps@NeurIPS 2022 as a full PMLR paper! :sparkle: arXiv, GitHub
Sep30, 2022 New Paper: Minimalistic Unsupervised Learning with the Sparse Manifold Transform. :wink: arXiv
Sep9, 2022
I gave a talk at Bay Area Vision Research Day (BAVRD) on the principles of unsupervised representation learning. :coffee:
Aug16, 2022
Compact and Optimal Deep Learning with Recurrent Parameter Generators, is accepted to WACV 2022! :sparkle: arXiv