MLPerf Training Working Group
Mission
Define, develop, and conduct the MLPerf Training benchmarks.
Purpose
Benchmarking the performance of training ML models on a wide variety of use cases, software, and hardware drives AI performance across the tech industry. The MLPerf Training working group draws on expertise in AI and the technology that powers AI from across the industry to design and create industry-standard benchmarks. Together, we create the reference implementations, rules, policies, and procedures to benchmark a wide variety of AI workloads.
The MLPerf Training working group strives for a critical balance of perspectives to ensure fairness and accuracy in the benchmarks. This balance comes from our members’ diverse experience in many different AI hardware and software spaces. We are always looking for new members to help us create the benchmarks that best capture innovation in AI.
Deliverables
- Training benchmark roadmap
- Training benchmark rules
- Training benchmark reference implementations
- Training benchmark results every ~6 months
Join
Meeting Schedule
Thursday May 16, 2024
Weekly – 08:35 – 10:00 Pacific Time
Results Publication
November 13, 2024
Wednesday
Related Blog
-
New MLPerf Training Benchmark Results Highlight Hardware and Software Innovations in AI Systems
Two new benchmarks added – highlighting language model fine-tuning and classification for graph data
-
LoRA selected as the fine-tuning technique added to MLPerf Training v4.0
MLPerf Training task force shares insights on the selection process for a new fine-tuning benchmark
-
Introducing the MLPerf Training Benchmark for Graph Neural Networks
Continued evolution to keep pace with advancements in AI
How to Join and Access MLPerf Training Working Group Resources
The MLPerf Training working group is limited exclusively to MLCommons members and affiliates. If you are not already a member or affiliate, or part of a member or affiliate company, you can learn more about MLCommons membership here.
- To sign up for the group mailing list, receive the meeting invite, and access shared documents and meeting minutes:
- Fill out our subscription form and indicate that you’d like to join the MLPerf Training Working Group.
- Associate a Google account with your organizational email address.
- Once your request to join the Training Working Group is approved, you’ll be able to access the Training folder in the Members Google Drive.
- To engage in working group discussions, join the group’s channels on the MLCommons Discord server.
- To access the GitHub repository (public):
- If you want to contribute code, please submit your GitHub ID to our subscription form.
- Visit the GitHub repository.
Training Working Group Chairs
To contact all MLCommons Training working group chairs email [email protected].
Hiwot Kassa
Hiwot is a research engineer at AI SW/HW codesign team at Meta working on performance optimization and benchmarking of large-scale workloads. She holds a Ph.D. in computer science and engineering from the University of Michigan.
Shriya Rishab
Shriya is a Senior Deep Learning Engineer at NVIDIA, and she works on benchmarking and scaling models while achieving state-of-the-art accuracy in the DL Algorithms team. Since 2021, she has contributed to new training benchmarks for MLPerf and has helped establish rules for the training suite. She holds a Master’s degree in Deep Learning from Columbia University and a Bachelor’s degree in Computer Science and Engineering from RV College of Engineering Bangalore in India.