LabelBench: A Comprehensive Framework for Benchmarking Adaptive Label-Efficient Learning
This provides a modular benchmark for researchers and practitioners to systematically compare and advance label-efficient learning methods, though it is incremental as it builds on existing techniques.
The paper tackles the lack of a comprehensive framework for evaluating combined label-efficient learning techniques by introducing LabelBench, a computationally-efficient framework that enables joint assessment of methods like active and semi-supervised learning, and applies it to show improved label-efficiencies in fine-tuning vision transformers.
Labeled data are critical to modern machine learning applications, but obtaining labels can be expensive. To mitigate this cost, machine learning methods, such as transfer learning, semi-supervised learning and active learning, aim to be label-efficient: achieving high predictive performance from relatively few labeled examples. While obtaining the best label-efficiency in practice often requires combinations of these techniques, existing benchmark and evaluation frameworks do not capture a concerted combination of all such techniques. This paper addresses this deficiency by introducing LabelBench, a new computationally-efficient framework for joint evaluation of multiple label-efficient learning techniques. As an application of LabelBench, we introduce a novel benchmark of state-of-the-art active learning methods in combination with semi-supervised learning for fine-tuning pretrained vision transformers. Our benchmark demonstrates better label-efficiencies than previously reported in active learning. LabelBench's modular codebase is open-sourced for the broader community to contribute label-efficient learning methods and benchmarks. The repository can be found at: https://github.com/EfficientTraining/LabelBench.