DataPerf: Benchmarks for Data-Centric AI Development
This work tackles the problem of data neglect in ML research, which causes real-world application issues, by providing a community-led benchmark for academia and industry, though it is incremental as it builds on existing data-centric concepts.
The paper introduces DataPerf, a benchmark suite for evaluating datasets and data-centric algorithms in machine learning, aiming to address issues like inaccuracy and bias by enabling iterative development on datasets across five initial benchmarks in vision, speech, and other modalities.
Machine learning research has long focused on models rather than datasets, and prominent datasets are used for common ML tasks without regard to the breadth, difficulty, and faithfulness of the underlying problems. Neglecting the fundamental importance of data has given rise to inaccuracy, bias, and fragility in real-world applications, and research is hindered by saturation across existing dataset benchmarks. In response, we present DataPerf, a community-led benchmark suite for evaluating ML datasets and data-centric algorithms. We aim to foster innovation in data-centric AI through competition, comparability, and reproducibility. We enable the ML community to iterate on datasets, instead of just architectures, and we provide an open, online platform with multiple rounds of challenges to support this iterative development. The first iteration of DataPerf contains five benchmarks covering a wide spectrum of data-centric techniques, tasks, and modalities in vision, speech, acquisition, debugging, and diffusion prompting, and we support hosting new contributed benchmarks from the community. The benchmarks, online evaluation platform, and baseline implementations are open source, and the MLCommons Association will maintain DataPerf to ensure long-term benefits to academia and industry.