An Exploration of How Training Set Composition Bias in Machine Learning Affects Identifying Rare Objects
This addresses a fundamental issue in classification for any application with imbalanced data, though the impact is described as modest and incremental.
The paper tackles the problem of training set composition bias in machine learning classifiers for rare objects, showing that common practices like up-weighting or balancing data can lead to over-assignment to the rare class, and it explores methods to detect and reduce this bias.
When training a machine learning classifier on data where one of the classes is intrinsically rare, the classifier will often assign too few sources to the rare class. To address this, it is common to up-weight the examples of the rare class to ensure it isn't ignored. It is also a frequent practice to train on restricted data where the balance of source types is closer to equal for the same reason. Here we show that these practices can bias the model toward over-assigning sources to the rare class. We also explore how to detect when training data bias has had a statistically significant impact on the trained model's predictions, and how to reduce the bias's impact. While the magnitude of the impact of the techniques developed here will vary with the details of the application, for most cases it should be modest. They are, however, universally applicable to every time a machine learning classification model is used, making them analogous to Bessel's correction to the sample variance.