Publications (by topics)

(* = equal contribution )
(** = alphabetical order )
(† = co-last author )
( Each paper appears in a single topic )

Graphical models / Hierarchical models / Mixture models / Bayesian nonparametrics (Theories and Applications in Machine Learning and Deep Learning)

• Is temperature sample efficient for softmax Gaussian mixture of experts? . Under review.
  Huy Nguyen, Pedram Akbarian, Nhat Ho.

• On least square estimation in softmax gating mixture of experts . Under review.
  Huy Nguyen, Nhat Ho†, Alessandro Rinaldo†.

• Bayesian nonparametrics meets data-driven robust optimization . Under review.
  Nicola Bariletto, Nhat Ho.

• FuseMoE: Mixture-of-experts Transformers for fleximodal fusion . Under review.
  Xing Han, Huy Nguyen, Carl William Harris, Nhat Ho†, Suchi Saria†.

• CompeteSMoE - Effective training of sparse mixture of experts via competition . Under review.
  Quang Pham, Truong Giang Do, Huy Nguyen, TrungTin Nguyen, Chenghao Liu, Mina Sartipi, Binh T. Nguyen, Savitha Ramasamy, Xiaoli Li, Steven Hoi†, Nhat Ho†.

• Demystifying softmax gating in Gaussian mixture of experts . Advances in NeurIPS, 2023 (Spotlight).
  Huy Nguyen, Tin Nguyen, Nhat Ho.

• Statistical perspective of top-K sparse softmax gating mixture of experts . ICLR, 2024.
  Huy Nguyen, Pedram Akbarian, Fanqi Yan, Nhat Ho.

• Minimax optimal rate for parameter estimation in multivariate deviated models. Advances in NeurIPS, 2023.
  Dat Do*, Huy Nguyen*, Khai Nguyen, Nhat Ho.

• Towards convergence rates for parameter estimation in Gaussian-gated mixture of experts . AISTATS, 2024.
  Huy Nguyen, Tin Nguyen, Khai Nguyen, Nhat Ho.

• On parameter estimation in deviated Gaussian mixture of experts. AISTATS, 2024.
  Huy Nguyen, Khai Nguyen, Nhat Ho.

• On excess mass behavior in Gaussian mixture models with Orlicz Wasserstein distances . Proceedings of the ICML, 2023.
  Aritra Guha, Nhat Ho, Long Nguyen.

• Convergence rates for Gaussian mixtures of experts. Journal of Machine Learning Research (JMLR), 2022.
  Nhat Ho, Chiao-Yu Yang, Michael I. Jordan.

• Beyond black box densities: Parameter learning for the deviated components. Advances in NeurIPS, 2022.
  Dat Do*, Nhat Ho*, Long Nguyen.

• Uniform convergence rates for maximum likelihood estimation under two-component Gaussian mixture models. Arxiv Preprint; Longer version to be submitted.
  Tudor Manole*, Nhat Ho*.

• Refined convergence rates for maximum likelihood estimation under finite mixture models. Proceedings of the ICML, 2022 (Long Presentation).
  Tudor Manole, Nhat Ho.

• On posterior contraction of parameters and interpretability in Bayesian mixture modeling . Bernoulli 27 (4), 2159-2188, 2021.
  Aritra Guha, Nhat Ho, XuanLong Nguyen.

• On posterior inference for the number of Clusters in Dirichlet process mixture models. Under review.
  Chiao-Yu Yang, Eric Xia, Nhat Ho, Michael I. Jordan.

• Weak separation in mixture models and implications for principal stratification. AISTATS, 2022.
  Nhat Ho**, Avi Feller**, Evan Greif**, Luke W. Miratrix**, Natesh S. Pillai**.

• Singularity structures and impacts on parameter estimation behavior in finite mixtures of distributions. SIAM Journal on Mathematics of Data Science (SIMODS), 1(4), 730–758, 2019.
  Nhat Ho and XuanLong Nguyen.

• Robust estimation of mixing measures in finite mixture models. Bernoulli, 26(2), 828-857, 2020.
  Nhat Ho, XuanLong Nguyen, Ya'acov Ritov.

• Convergence rates of parameter estimation for some weakly identifiable finite mixtures. Annals of Statistics, 44(6), 2726-2755, 2016.
  Nhat Ho and XuanLong Nguyen.

• On strong identifiability and convergence rates of parameter estimation in finite mixtures. Electronic Journal of Statistics, 10(1), 271-307, 2016.
  Nhat Ho and XuanLong Nguyen.

Optimal transport (Computation and Methods in Statistical Machine Learning and Deep Learning)

• On the efficiency of entropic regularized algorithms for optimal transport . Journal of Machine Learning Research (JMLR), 2022.
  Tianyi Lin, Nhat Ho, Michael I. Jordan.

• Sliced Wasserstein with random-path projecting directions . Under review.
  Khai Nguyen, Shujian Zhang, Tam Le, Nhat Ho.

• On the complexity of approximating multi-marginal optimal transport. Journal of Machine Learning Research (JMLR), 2022.
  Tianyi Lin*, Nhat Ho*, Marco Cuturi, Michael I. Jordan.

• Energy-based sliced Wasserstein distance . Advances in NeurIPS, 2023.
  Khai Nguyen, Nhat Ho.

• Markovian sliced Wasserstein distances: Beyond independent projections . Advances in NeurIPS, 2023.
  Khai Nguyen, Tongzheng Ren, Nhat Ho.

• Quasi-Monte Carlo for 3D sliced Wasserstein . Under review.
  Khai Nguyen, Nicola Bariletto, Nhat Ho.

• Control variate sliced Wasserstein estimators . Under review.
  Khai Nguyen, Nhat Ho.

• Diffeomorphic deformation via sliced Wasserstein distance optimization for cortical surface reconstruction. Under review.
  Tung Le, Khai Nguyen, Shanlin Sun, Kun Han, Nhat Ho, Xiaohui Xie.

• Hierarchical sliced Wasserstein distance. ICLR, 2023.
  Khai Nguyen, Tongzheng Ren, Huy Nguyen, Litu Rout, Tan Nguyen, Nhat Ho.

• Self-attention amortized distributional projection optimization for sliced Wasserstein point-clouds reconstruction . Proceedings of the ICML, 2023.
  Khai Nguyen*, Dang Nguyen*, Nhat Ho.

• Revisiting projected Wasserstein metric on images: from vectorization to convolution. Advances in NeurIPS, 2022.
  Khai Nguyen, Nhat Ho.

• Amortized projection optimization for sliced Wasserstein generative models. Advances in NeurIPS, 2022.
  Khai Nguyen, Nhat Ho.

• Improving minibatch optimal transport via partial transportation. Proceedings of the ICML, 2022.
  Khai Nguyen, Dang Nguyen, Anh Le, Tung Pham, Nhat Ho.

• Entropic Gromov-Wasserstein between Gaussian distributions. Proceedings of the ICML, 2022.
  Khang Le*, Dung Le*, Huy Nguyen*, Dat Do, Tung Pham, Nhat Ho.

• BoMb-OT: On batch of mini-batches optimal transport . Proceedings of the ICML, 2022.
  Khai Nguyen, Dang Nguyen, Quoc Nguyen, Tung Pham, Dinh Phung, Hung Bui, Trung Le, Nhat Ho.

• On multimarginal partial optimal transport: Equivalent forms and computational complexity. AISTATS, 2022.
  Huy Nguyen*, Khang Le*, Tung Pham, Nhat Ho.

• On robust optimal transport: Computational complexity and barycenter computation . Advances in NeurIPS, 2021.
  Huy Nguyen*, Khang Le*, Quang Nguyen, Tung Pham, Hung Bui*, Nhat Ho*.

• Fixed-support Wasserstein barycenters: computational hardness and fast algorithm . Advances in NeurIPS, 2020.
  Tianyi Lin, Nhat Ho, Xi Chen, Marco Cuturi, Michael I. Jordan.

• Projection robust Wasserstein distance and Riemannian optimization . Advances in NeurIPS, 2020 (Spotlight).
  Tianyi Lin*, Chenyou Fan*, Nhat Ho, Marco Cuturi, Michael I. Jordan.

• Flow-based alignment approaches for probability measures in different spaces. AISTATS, 2021.
  Tam Le*, Nhat Ho*, Makoto Yamada.

• On scalable variant of Wasserstein barycenter . Under review.
  Tam Le*, Viet Huynh*, Nhat Ho*, Dinh Phung, Makoto Yamada.

• On unbalanced optimal transport: an analysis of Sinkhorn algorithm. Proceedings of the ICML, 2020.
  Khiem Pham*, Khang Le*, Nhat Ho, Tung Pham, Hung Bui.

• Fast algorithms for computational optimal transport and Wasserstein barycenter. AISTATS, 2020.
  Wenshuo Guo, Nhat Ho, Michael I. Jordan.

• On efficient optimal transport: an analysis of greedy and accelerated mirror descent algorithms. Proceedings of the ICML, 2019.
  Tianyi Lin*, Nhat Ho*, Michael I. Jordan.

• Probabilistic multilevel clustering via composite transportation distance. AISTATS, 2019.
  Nhat Ho*, Viet Huynh*, Dinh Phung, Michael I. Jordan.

• Multilevel clustering via Wasserstein means. Proceedings of the ICML, 2017.
  Nhat Ho, XuanLong Nguyen, Mikhail Yurochkin, Hung Hai Bui, Viet Huynh, and Dinh Phung.
  
  Journal extension: On efficient multilevel clustering via Wasserstein distances. Journal of Machine Learning Research (JMLR), 22, 1-43, 2021.

Optimization in statistical settings / Distributed computing

• Instability, computational efficiency, and statistical accuracy . Accepted with minor revision at Journal of Machine Learning Research (JMLR), 2023.
  Nhat Ho*, Raaz Dwivedi*, Koulik Khamaru*, Martin J. Wainwright, Michael I. Jordan, Bin Yu.

• On the computational and statistical complexity of over-parameterized matrix sensing . Under review.
  Jiacheng Zhuo, Jeongyeol Kwon, Nhat Ho, Constantine Caramanis.

• Beyond EM algorithm on over-specified two-component location-scale Gaussian mixtures. Under review.
  Tongzheng Ren*, Fuheng Cui*, Sujay Sanghavi, Nhat Ho.

• An exponentially increasing step-size for parameter estimation in statistical models. Under review.
  Nhat Ho**, Tongzheng Ren**, Purnamrita Sarkar**, Sujay Sanghavi**, Rachel Ward**.

• Improving computational complexity in statistical models with second-order information . Under review.
  Tongzheng Ren, Jiacheng Zhuo, Sujay Sanghavi, Nhat Ho.

• Statistical and computational complexities of BFGS quasi-Newton method for generalized linear models. Under review.
  Qiujiang Jin, Tongzheng Ren, Nhat Ho, Aryan Mokhtari.

• Probabilistic best subset selection via gradient-based optimization. Under review.
  Mingzhang Yin, Nhat Ho, Bowei Yan, Xiaoning Qian, Mingyuan Zhou.

• Towards statistical and computational complexities of Polyak step size gradient descent . AISTATS, 2022.
  Tongzheng Ren*, Fuheng Cui*, Alexia Atsidakou*, Sujay Sanghavi, Nhat Ho.

• On the minimax optimality of the EM algorithm for learning two-component mixed linear regression. AISTATS, 2021.
  Jeong Y. Kwon, Nhat Ho, Constantine Caramanis.

• On structured filtering-clustering: Global error bound and optimal first-order algorithms. AISTATS, 2022.
  Nhat Ho*, Tianyi Lin*, Michael I. Jordan.

• Singularity, misspecification, and the convergence rate of EM. Annals of Statistics, 48(6), 3161-3182, 2020.
  Raaz Dwivedi*, Nhat Ho*, Koulik Khamaru*, Martin J. Wainwright, Michael I. Jordan, Bin Yu.

• Sharp analysis of Expectation-Maximization for weakly identifiable models. AISTATS, 2020.
  Raaz Dwivedi*, Nhat Ho*, Koulik Khamaru*, Martin J. Wainwright, Michael I. Jordan, Bin Yu.

• Theoretical guarantees for EM under misspecified Gaussian mixture models . Advances in NeurIPS 31, 2018.
  Raaz Dwivedi*, Nhat Ho*, Koulik Khamaru*, Martin J. Wainwright, Michael I. Jordan.

Sampling and Markov chains / (Approximate) Bayesian inference

• Bayesian consistency with the supremum metric . Statistica Sinica, 2022.
  Nhat Ho**, Stephen G. Walker**.

• A diffusion process perspective on the posterior contraction rates for parameters. SIAM Journal on Mathematics of Data Science (SIMODS), 2023.
  Wenlong Mou*, Nhat Ho*, Martin J. Wainwright, Peter L. Bartlett, Michael I. Jordan.

• Sampling for Bayesian mixture models: MCMC with polynomial-time mixing. Under revision.
  Wenlong Mou, Nhat Ho, Martin J. Wainwright, Peter L. Bartlett, Michael I. Jordan.

• A partial differential equation perspective on Berstein-Von Mises theorem. To be submitted.
  Wenlong Mou, Nhat Ho, Martin J. Wainwright, Peter L. Bartlett, Michael I. Jordan.

Nonparametric estimation

• Multivariate smoothing via the Fourier integral theorem and Fourier kernel . Under review.
  Nhat Ho**, Stephen G. Walker**.

• Statistical analysis using the Fourier integral theorem . Under review.
  Nhat Ho**, Stephen G. Walker**.

• On integral theorems: Monte Carlo estimators and optimal functions . Under review.
  Nhat Ho**, Stephen G. Walker**.

• BONuS: Multiple multivariate testing with a data-adaptive test statistic . Under review.
  Chiao-Yu Yang, Lihua Lei, Nhat Ho, Will Fithian.

Deep learning (Generative models, Transformer, Transfer learning, Federated Learning, Bayesian neural networks, etc.)

• A Bayesian perspective of convolutional neural networks through a deconvolutional generative model. Journal of Machine Learning Research, 2023.
  Nhat Ho*, Tan Nguyen*, Ankit Patel, Anima Anandkumar, Michael I. Jordan, Richard Baraniuk.

• A primal-dual framework for transformers and neural networks. ICLR, 2023 (Spotlight).
  Tan Minh Nguyen, Tam Minh Nguyen, Nhat Ho, Andrea L. Bertozzi, Richard Baraniuk, Stanley Osher.

• Neural collapse in deep linear network: from balanced to imbalanced data . Proceedings of the ICML, 2023.
  Hien Dang*, Tan Nguyen*, Tho Tran, Hung Tran†, Nhat Ho†.

• Revisiting over-smoothing and over-squashing using Ollivier's Ricci curvature . Proceedings of the ICML, 2023.
  Khang Nguyen, Tan Minh Nguyen, Nong Minh Hieu, Vinh Duc Nguyen, Nhat Ho†, Stanley Osher†.

• LVM-Med: Learning large-Scale self-Supervised vision models for medical imaging via second-order graph matching. Advances in NeurIPS, 2023.
  Duy Nguyen, Hoang Nguyen, Nghiem Diep, Tan Pham, Tri Cao, Binh T. Nguyen, Paul Swoboda, Nhat Ho, Shadi Albarqouni, Pengtao Xie, Mathias Niepert, Daniel Sonntag.

• Robustify Transformers with robust kernel density estimation. Advances in NeurIPS, 2023.
  Xing Han*, Tongzheng Ren*, Tan Minh Nguyen*, Khai Nguyen, Joydeep Ghosh, Nhat Ho.

• Privacy preserving Bayesian federated learning in heterogeneous settings. Under review.
  Disha Makhija, Nhat Ho†, Joydeep Ghosh†.

• Improving multi-task learning via seeking task-based flat regions. Under review.
  Hoang Phan, Tung Tran, Ngoc Tran, Nhat Ho, Dinh Phung, Trung Le.

• Model fusion of heterogeneous neural networks via cross-layer alignment . IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2023 (Top 3%).
  Dang Nguyen, Trang Nguyen, Khai Nguyen, Dinh Phung, Hung Bui, Nhat Ho.

• A probabilistic framework for pruning transformers via a finite admixture of keys . IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2023 (Top 3%).
  Tam Nguyen*, Tan Nguyen*, Long Bui, Hai Do, Dung Le, Hung Tran-The, Khuong Nguyen, Nhat Ho, Stanley Osher, Richard Baraniuk.

• FourierFormer: Transformer meets generalized Fourier integral attentions . Advances in NeurIPS, 2022.
  Tan Nguyen*, Minh Pham*, Tam Nguyen, Khai Nguyen, Stanley Osher, Nhat Ho.

• Improving Transformer with an admixture of attention heads . Advances in NeurIPS, 2022.
  Tam Nguyen*, Tan Nguyen*, Hai Do, Khai Nguyen, Vishwanath Saragadam, Minh Pham, Khuong Nguyen,Stanley Osher†, Nhat Ho†.

• Federated self-supervised learning for heterogeneous clients. Under review.
  Disha Makhija, Nhat Ho†, Joydeep Ghosh†.

• Global-local regularization via distributional robustness. AISTATS, 2023.
  Hoang Phan, Trung Le, Trung Phung, Tuan Anh Bui, Nhat Ho, Dinh Phung.

• Stochastic multiple target sampling gradient descent . Advances in NeurIPS, 2022.
  Hoang Phan, Ngoc Tran, Trung Le, Toan Tran, Nhat Ho, Dinh Phung.

• Improving Transformers with probabilistic attention keys . Proceedings of the ICML, 2022.
  Tam Nguyen*, Tan Nguyen*, Dung Le, Khuong Nguyen, Anh Tran, Richard Baraniuk, Stanley Osher†, Nhat Ho†.

• Architecture agnostic federated learning for neural networks. Proceedings of the ICML, 2022.
  Disha Makhija, Xing Han, Nhat Ho, Joydeep Ghosh.

• LAMDA: Label matching deep domain adaptation. Proceedings of the ICML, 2021.
  Trung Le, Tuan Nguyen, Nhat Ho, Hung Bui, Dinh Phung.

• Distributional sliced-Wasserstein and applications to deep generative modeling. ICLR, 2021.
  Khai Nguyen, Nhat Ho, Tung Pham, Hung Bui.

• Point-set distances for learning representations of 3D point clouds. International Conference on Computer Vision (ICCV), 2021.
  Trung Nguyen, Hieu Pham, Tam Le, Nhat Ho, Tung Pham, Son Hua.

• Structured Dropout variational inference for Bayesian neural networks . Advances in NeurIPS, 2021.
  Son Nguyen*, Duong Nguyen*, Khai Nguyen, Khoat Than, Hung Bui†, Nhat Ho†.

• Improving relational regularized autoencoders with spherical sliced fused Gromov Wasserstein. ICLR, 2021.
  Khai Nguyen, Son Nguyen, Nhat Ho, Tung Pham, Hung Bui.