News
- 03/10/2024: Our paper “FSMoE: A Flexible and Scalable Training System for Sparse Mixture-of-Experts Models” has been accepted by ASPLOS 2025.
- 20/09/2024: I am awarded as one of Outstanding Contributors to Kunpeng&Shengteng Research Innovation (昇腾科研创新卓越贡献者).
- 16/09/2024: I has been ranked 2024 World’s Top 2% Scientists released by Stanford University.
- 11/06/2024: Two papers have been accepted by ICPP 2024.
- 15/04/2024: Our paper “Scheduling Deep Learning Jobs in Multi-Tenant GPU Clusters via Wise Resource Sharing” has been accepted by IEEE/ACM IWQoS 2024.
- 16/01/2024: Our paper “FedImpro: Measuring and Improving Client Update in Federated Learning” has been accepted by ICLR 2024.
- 10/01/2024: Our paper “ScheMoE: An Extensible Mixture-of-Experts Distributed Training System with Tasks Scheduling” has been accepted by EuroSys 2024.
- 01/12/2023: Our paper “Parm: Efficient Training of Large Sparsely-Activated Models with Dedicated Schedules” has been accepted by IEEE INFOCOM 2024.
- 12/11/2023: Our paper “Performance Analysis and Optimizations of Matrix Multiplications on ARMv8 Processors” has been accepted by DATE 2024.
- 10/04/2023: Two papers have been accepted by IEEE ICDCS 2023.
- 21/01/2023: Our paper “Eva: Practical Second-order Optimization with Kronecker-vectorized Approximation” has been accepted by ICLR 2023.
- 06/12/2022: Our paper “GossipFL: A Decentralized Federated Learning Framework with Sparsified and Adaptive Communication” has been accepted by IEEE TPDS.
- 02/12/2022: Two papers have been accepted by IEEE INFOCOM 2023.
- 06/09/2022: Our paper “Scalable K-FAC Training for Deep Neural Networks with Distributed Preconditioning” has been accepted by IEEE Transactions on Cloud Computing.
- 01/09/2022: I joined Harbin Institute of Technology, Shenzhen as an Assistant Professor.
- 04/07/2022: Our paper “EASNet:Searching Elastic and Accurate Network Architecture for Stereo Matching” has been accepted by ECCV 2022.
- 15/05/2022: Our paper “Virtual Homogeneity Learning: Defending against Data Heterogeneity in Federated Learning” has been accepted by ICML 2022.
- 11/05/2021: Our paper “Exploiting Simultaneous Communications to Accelerate Data Parallel Distributed Deep Learning” has recieved the Best Paper Award (3 out of 252 accepted papers) by IEEE INFOCOM 2021.
- 18/03/2021: Our paper “Accelerating Distributed K-FAC with Smart Parallelism of Computing and Communication Tasks” has been accepted by IEEE ICDCS 2021.
- 17/03/2021: We are organizing the 5th International Workshop on Embedded and Mobile Deep Learning: https://emdl21.github.io/, co-located with ACM MobiSys 2021. Please consider submitting your work to the venue.
- 19/01/2021: Our paper “Towards Scalable Distributed Training of Deep Learning on Public Cloud Clusters” [PDF] has been accepted by MLSys 2021.
- 14/01/2021: Our paper “MG-WFBP: Merging Gradients Wisely for Efficient Communication in Distributed Deep Learning” [Code, PDF] has been accepted by IEEE TPDS.
- 05/12/2020: Our paper “Exploiting Simultaneous Communications to Accelerate Data Parallel Distributed Deep Learning” has been accepted by IEEE INFOCOM 2021.
- 02/12/2020: Our paper “Automated Model Design and Benchmarking of 3D Deep Learning Models for COVID-19 Detection with Chest CT Scans” has been accepted by AAAI 2021.
- 21/10/2020: Our survey paper “A Quantitative Survey of Communication Optimizations in Distributed Deep Learning” [Code, PDF] has been accepted by IEEE Network Magazine.
- 28/09/2020: Awarded CCF-Baidu Open Fund (CCF-百度松果基金) [Link].
- 01/09/2020: I start my academic position as a RAP at HKUST.