Leveraging Importance Weights in Subset Selection
This work addresses subset selection for machine learning practitioners needing efficient batch processing, but it is incremental as it builds on existing importance sampling and entropy-based methods.
The paper tackled the problem of subset selection in a practical batch setting where examples are sampled one at a time but model updates occur only after collecting a batch, by introducing IWeS, an algorithm that uses importance sampling based on model entropy from previous batches, resulting in significant performance improvements on seven datasets and competitive results in active learning.
We present a subset selection algorithm designed to work with arbitrary model families in a practical batch setting. In such a setting, an algorithm can sample examples one at a time but, in order to limit overhead costs, is only able to update its state (i.e. further train model weights) once a large enough batch of examples is selected. Our algorithm, IWeS, selects examples by importance sampling where the sampling probability assigned to each example is based on the entropy of models trained on previously selected batches. IWeS admits significant performance improvement compared to other subset selection algorithms for seven publicly available datasets. Additionally, it is competitive in an active learning setting, where the label information is not available at selection time. We also provide an initial theoretical analysis to support our importance weighting approach, proving generalization and sampling rate bounds.