Research

Visit Google Scholar for a full list of publications

Recent papers by topic

High dimensional isoperimetry and related

• Digesting the proof of the sharp thin-shell inequality [arXiv], Yuansi Chen, Boaz Klartag. Submitted to arXiv on July 25, 2026
  The proof was found using GPT Pro 5.6 and it was checked by the two authors. A preprint appeared on Klartag's website on July 25. And a preliminary version of the draft appeared on a public Github repository on July 19 for a short time and the repository had over 70 git clones during that short time.
• Talagrand's convolution conjecture up to loglog via perturbed reverse heat [arXiv], Yuansi Chen
• An almost constant lower bound of the isoperimetric coefficient in the KLS conjecture [GAFA] [arXiv], Yuansi Chen. GAFA 2021

Markov chain Monte Carlo methods/convergence analysis

• Regularized Dikin Walks for Sampling Truncated Logconcave Measures, Mixed Isoperimetry and Beyond Worst-Case Analysis, [arXiv], Minhui Jiang, Yuansi Chen. COLT 2025
• When does Metropolized Hamiltonian Monte Carlo provably outperform Metropolis-adjusted Langevin algorithm?, [arXiv], Yuansi Chen, Khashayar Gatmiry
• Hit-and-run mixing via localization schemes, [arXiv], Yuansi Chen, Ronen Eldan
• Localization Schemes: A Framework for Proving Mixing Bounds for Markov Chains, [DMJ] [arXiv], Yuansi Chen, Ronen Eldan. Duke Math. J. 2025
• Minimax Mixing Time of the Metropolis-Adjusted Langevin Algorithm for Log-Concave Sampling, [JMLR] [arXiv], Keru Wu, Scott Schmidler, Yuansi Chen. JMLR 2022

Domain adaptation/Transfer learning

• When few labeled target data suffice: a theory of semi-supervised domain adaptation via fine-tuning from multiple adaptive starts, [arXiv], Wooseok Ha, Yuansi Chen
• Prominent roles of conditionally invariant components in domain adaptation: Theory and algorithms, [JMLR] [arXiv], Keru Wu, Yuansi Chen, Wooseok Ha, Bin Yu. JMLR 2024
• Domain adaptation under structural causal models, [JMLR] [arXiv], Yuansi Chen, Peter Bühlmann. JMLR 2021

Papers with code

• Log-concave sampling: Metropolis-Hastings algorithms are fast, [JMLR][arXiv] [Code on Github], Yuansi Chen, Raaz Dwivedi, Martin Wainwright, Bin Yu. JMLR 2019
• Fast MCMC Algorithms on Polytopes, [JMLR] [arXiv][C++ Implementation on Github], Yuansi Chen, Raaz Dwivedi, Martin Wainwright, Bin Yu. JMLR 2018
• Fast and Robust Archetypal Analysis for Representation Learning, [CVPR] [arXiv][Code&Demo], Yuansi Chen, Julien Mairal and Zaid Harchaoui. CVPR 2014