LINs Lab

Welcome to the website of the Learning and INference Systems (LINs) Lab at Westlake University!

The research interests of the lab lie in the intersection of optimization and generalization for deep learning:

  • leveraging theoretical/empirical understanding (e.g., loss landscape, and training dynamics)
  • to design efficient & robust methods (both learning and inference)
  • for deep learning (centralized) and collaborative deep learning (distributed and/or decentralized),
  • under imperfect environments (e.g., noisy, heterogeneous, and hardware-constrained).

News

Feb 19, 2023 We are looking for a research assistant working on large-scale distributed deep learning training!
Feb 16, 2023 Congrats on Peng SUN (our undergraduate intern) for his first two first-author papers at ICASSP 2023.
Jan 29, 2023 One paper was accepted to ICLR 2023.
Dec 3, 2022 Our research seminar website is live.
Oct 18, 2022 Open positions: We have 1 joint postdoc position with Dr. Sebastian U. Stich.
Jun 12, 2022 Open positions: We have several open positions for 1 postdoc researcher, 2-3 Ph.D. students (for Fall 2023), and multiple research assistants/interns.

Selected publications

  1. ICLR 2023
    Test-Time Robust Personalization for Federated Learning
    Jiang, Liangze*, and Lin, Tao*†
    In International Conference on Learning Representations (ICLR) 2023
    PDF Code
  2. ICML 2021
    Quasi-global Momentum: Accelerating Decentralized Deep Learning on Heterogeneous Data
    Lin, Tao, Karimireddy, Sai Praneeth, Stich, Sebastian U., and Jaggi, Martin
    In International Conference on Machine Learning (ICML) 2021
    PDF Code
  3. NeurIPS 2020
    Ensemble Distillation for Robust Model Fusion in Federated Learning
    Lin, Tao*†, Kong, Lingjing*, Stich, Sebastian U, and Jaggi, Martin
    In Advances in Neural Information Processing Systems (NeurIPS) 2020
    PDF Code
  4. ICLR 2020
    Decentralized Deep Learning with Arbitrary Communication Compression
    Koloskova, Anastasia*, Lin, Tao*, Stich, Sebastian U, and Jaggi, Martin
    In International Conference on Learning Representations (ICLR) 2020
    PDF Code
  5. ICLR 2020
    Don’t Use Large Mini-batches, Use Local SGD
    Lin, Tao, Stich, Sebastian U, Patel, Kumar Kshitij, and Jaggi, Martin
    In International Conference on Learning Representations (ICLR) 2020
    PDF Code