Multiscale Feature Attribution for Outliers
This addresses the need for interpretable outlier analysis in domains like astronomy, though it is incremental as it builds on existing attribution methods.
The paper tackled the problem of interpreting which features make outliers anomalous, proposing Inverse Multiscale Occlusion for feature attribution on outliers where model performance is uncertain. The method was demonstrated on galaxy spectra outliers, showing results more interpretable than alternatives.
Machine learning techniques can automatically identify outliers in massive datasets, much faster and more reproducible than human inspection ever could. But finding such outliers immediately leads to the question: which features render this input anomalous? We propose a new feature attribution method, Inverse Multiscale Occlusion, that is specifically designed for outliers, for which we have little knowledge of the type of features we want to identify and expect that the model performance is questionable because anomalous test data likely exceed the limits of the training data. We demonstrate our method on outliers detected in galaxy spectra from the Dark Energy Survey Instrument and find its results to be much more interpretable than alternative attribution approaches.