swMATH ID: 33573
Software Authors: Andreas Kirsch, Joost van Amersfoort, Yarin Gal
Description: BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learning. We develop BatchBALD, a tractable approximation to the mutual information between a batch of points and model parameters, which we use as an acquisition function to select multiple informative points jointly for the task of deep Bayesian active learning. BatchBALD is a greedy linear-time 1−1e-approximate algorithm amenable to dynamic programming and efficient caching. We compare BatchBALD to the commonly used approach for batch data acquisition and find that the current approach acquires similar and redundant points, sometimes performing worse than randomly acquiring data. We finish by showing that, using BatchBALD to consider dependencies within an acquisition batch, we achieve new state of the art performance on standard benchmarks, providing substantial data efficiency improvements in batch acquisition.
Homepage: https://arxiv.org/abs/1906.08158
Source Code:  https://github.com/BlackHC/BatchBALD
Related Software: ImageNet; PyTorch; Cardinal; modAL; PyTorch Lightning; SuperGLUE; UCI-ml; Scikit; libact; ALiPy; scikit-activeml; JCLAL; skorch; GPyTorch; Flax; JAX; Keras; Python; PyRelationAL; NumPy
Cited in: 1 Document

Cited by 1 Author

1 Hino, Hideitsu

Cited in 1 Field

1 Statistics (62-XX)

Citations by Year