publications

(*) denotes equal contribution.

For a complete list, visit my Google Scholar profile.

2025

  1. Preprint
    Sigmoid Self-Attention is Better than Softmax Self-Attention: A Mixture-of-Experts Perspective
    Fanqi Yan, Huy Nguyen, Pedram Akbarian, Nhat Ho, and Alessandro Rinaldo
    arXiv:2502.00281, 2025
    Under review
    arXiv
  2. AISTATS
    Understanding Expert Structures on Minimax Parameter Estimation in Contaminated Mixture of Experts
    Fanqi Yan, Huy Nguyen, Le Quang Dung, Pedram Akbarian, and Nhat Ho
    In International Conference on Artificial Intelligence and Statistics (AISTATS), 2025
    arXiv
  3. ICLR
    Statistical Advantages of Perturbing Cosine Router in Sparse Mixture of Experts
    Huy Nguyen, Pedram Akbarian*, Trang Pham*, Trang Nguyen, Shujian Zhang, and Nhat Ho
    In International Conference on Learning Representations (ICLR), 2025
    arXiv

2024

  1. Preprint
    Quadratic Gating Functions in Mixture of Experts: A Statistical Insight
    Pedram Akbarian*, Huy Nguyen*, Xing Han*, and Nhat Ho
    arXiv:2410.11222, 2024
    Under review
    arXiv
  2. ICML
    Improving Computational Complexity in Statistical Models with Local Curvature Information
    Pedram Akbarian*, Tongzheng Ren*, Jiacheng Zhuo, Nhat Ho, and  others
    In International Conference on Machine Learning (ICML), 2024
    PDF
  3. ICML
    Is Temperature Sample Efficient for Softmax Gaussian Mixture of Experts?
    Huy Nguyen, Pedram Akbarian, and Nhat Ho
    In International Conference on Machine Learning (ICML), 2024
    arXiv
  4. ICML
    A General Theory for Softmax Gating Multinomial Logistic Mixture of Experts
    Huy Nguyen, Pedram Akbarian, TrungTin Nguyen, and Nhat Ho
    In International Conference on Machine Learning (ICML), 2024
    arXiv
  5. ICLR
    Statistical Perspective of Top-K Sparse Softmax Gating Mixture of Experts
    Huy Nguyen, Pedram Akbarian, Fanqi Yan, and Nhat Ho
    In International Conference on Learning Representations (ICLR), 2024
    arXiv

2022

  1. NeurIPS
    Improving Counterfactual Explanations for Time Series Classification Models in Healthcare Settings
    Tina Han, Jette Henderson, Pedram Akbarian, and Joydeep Ghosh
    In NeurIPS 2022 Workshop on Learning from Time Series for Health, 2022
    PDF