Talks and Poster Presentations

2025 Talks and Poster Presentations

  1. Rio Othon Palace
    July 21-25, 2025
    Presented a poster on MindFlayer SGD: Efficient Parallel SGD
    in the Presence of Heterogeneous and Random Worker Compute Times
  2. Vancouver Convention Center
    July 13-19, 2025
    Presented posters on
    • Ringmaster ASGD: The First Asynchronous SGD with Optimal Time Complexity
    • ATA: Adaptive Task Allocation for Efficient Resource Management
      in Distributed Machine Learning
  3. KAUST
    May 18, 2025
    Presented a poster on ATA: Adaptive Task Allocation for Efficient Resource Management
    in Distributed Machine Learning     [poster]
  4. Singapore EXPO
    April 24-28, 2025
    Presented a poster on LoCoDL: Communication-Efficient Distributed Learning
    with Local Training and Compression     Spotlight presentation (top 5.1% of the submitted papers)
  5. FLOW Talk #126
    April 16, 2025
    Delivered a talk on Ringmaster ASGD: The First Asynchronous SGD
    with Optimal Time Complexity     [video] [slides]
  6. KAUST
    April 7-10, 2025
    Presented a poster on ATA: Adaptive Task Allocation
    for Efficient Resource Management in Distributed Machine Learning     [poster]
  7. YSU Krisp-AI Lab
    March 28, 2025
    Delivered a talk on Ringmaster ASGD: The First Asynchronous SGD
    with Optimal Time Complexity     [slides]
  8. London, England
    King's House
    March 26, 2025
    Delivered a talk on Ringmaster ASGD: The First Asynchronous SGD
    with Optimal Time Complexity     [slides]
  9. Beihang University (BUAA)
    March 17, 2025
    Invited by Jiaxin Xie to give a talk on Ringmaster ASGD: The First Asynchronous SGD
    with Optimal Time Complexity     [slides]
  10. Beijing, China
    Beijing Institute of Mathematical Sciences and Applications (BIMSA)
    March 13, 2025
    Invited by Yi-Shuai Niu to give a talk on Ringmaster ASGD: The First Asynchronous SGD
    with Optimal Time Complexity     [slides]
  11. Beijing, China
    Peking University
    March 12, 2025
    Invited by Kun Yuan to give a talk on Ringmaster ASGD: The First Asynchronous SGD
    with Optimal Time Complexity     [slides]
  12. KAUST
    February 27, 2025
    Delivered a talk on Ringmaster ASGD: The First Asynchronous SGD
    with Optimal Time Complexity     [slides]

2024 Talks and Poster Presentations

  1. Vancouver Convention Center
    December 15, 2024
    Presented
    • MindFlayer: Efficient Asynchronous Parallel SGD in the Presence of Heterogeneous
      and Random Worker Compute Times       (Oral, Top 5%) [video]
    • Differentially Private Random Block Coordinate Descent
    • LoCoDL: Communication-Efficient Distributed Learning with Local Training and Compression
  2. MLR Weekly Seminar
    Online
    Machine Learning Research at Apple
    November 21, 2024
    Invited by Samy Bengio to give a talk on MindFlayer: Efficient Asynchronous Parallel SGD
    in the Presence of Heterogeneous and Random Worker Compute Times     [slides]
  3. Yerevan State University
    October 18, 2024
    Delivered a talk on MindFlayer: Efficient Asynchronous Parallel SGD
    in the Presence of Heterogeneous and Random Worker Compute Times
  4. KAUST, Saudi Arabia
    KAUST
    October 10, 2024
    Presented a poster on GradSkip: Communication-Accelerated Local Gradient Methods
    with Better Computational Complexity     [3rd place]
  5. Yerevan State University
    July 6, 2024
    Delivered a talk on MindFlayer: Efficient Asynchronous Parallel SGD in the Presence of Heterogeneous
    and Random Worker Compute Times     [abstract]
  6. KAUST
    May 27, 2024
    Presented a poster on GradSkip: Communication-Accelerated Local Gradient Methods
    with Better Computational Complexity     [poster]
  7. CS 331, Stochastic Gradient Descent Methods
    KAUST, Saudi Arabia
    KAUST
    May 5, 2024
    Delivered a guest lecture on MindFlayer: Efficient Asynchronous Parallel SGD
    in the Presence of Heterogeneous and Random Worker Compute Times
  8. Okinawa Institute of Science and Technology (OIST)
    March 13, 2024
    Presented a poster on GradSkip: Communication-Accelerated Local Gradient Methods
    with Better Computational Complexity     [poster]
  9. KAUST
    February 21, 2024
    Presented a poster on GradSkip: Communication-Accelerated Local Gradient Methods
    with Better Computational Complexity     [poster]

2023 Talks and Poster Presentations

  1. Group Seminar
    KAUST, Saudi Arabia
    KAUST
    November 16, 2023
    Delivered a talk on Differentially Private Coordinate Descent for Composite Empirical Risk Minimization
  2. University of Warwick
    October 6, 2023
    Delivered a talk on GradSkip: Communication-Accelerated Local Gradient Methods
    with Better Computational Complexity     [slides]
  3. Yerevan State University
    July 5, 2023
    Delivered a talk on GradSkip: Communication-Accelerated Local Gradient Methods
    with Better Computational Complexity     [abstract]
  4. Yerevan State University
    March 10, 2023
    Delivered a talk on GradSkip: Communication-Accelerated Local Gradient Methods
    with Better Computational Complexity     [video (Armenian)]

2022 Talks and Poster Presentations

  1. FLOW Talk #88
    December 7, 2022
    Delivered a talk on GradSkip: Communication-Accelerated Local Gradient Methods
    with Better Computational Complexity     [video]
  2. Yerevan State University
    April 10, 2022
    Delivered a talk on ProxSkip: Yes! Local Gradient Steps Provably Lead to Communication Acceleration! Finally!