Scope of ML based benchmarks in MLPerf.

[rakshithgb-fujitsu]

Over the years, MLCommons has primarily concentrated on deep learning. However, many real-world applications continue to rely on traditional machine learning algorithms such as k-means clustering, Support Vector Machines (SVM), and Logistic Regression etc.

The existing public benchmarks for machine learning workloads are outdated and poorly maintained. MLCommons has the opportunity to standardize these workflows and incorporate them into the MLPerf benchmarks.

As a starting point, Dataperf already includes several machine learning workflows, such as the 2023 speech selection task. These workflows could be standardized and integrated into MLPerf. I am interested in your thoughts on this proposal.