Shortcut Detection with Variational Autoencoders
This addresses the challenge of ensuring models generalize well in real-world ML applications, though it appears incremental as it builds on existing VAE methods for a known bottleneck.
The paper tackles the problem of detecting spurious correlations (shortcuts) in image and audio datasets using variational autoencoders, identifying previously undiscovered shortcuts in real-world datasets.
For real-world applications of machine learning (ML), it is essential that models make predictions based on well-generalizing features rather than spurious correlations in the data. The identification of such spurious correlations, also known as shortcuts, is a challenging problem and has so far been scarcely addressed. In this work, we present a novel approach to detect shortcuts in image and audio datasets by leveraging variational autoencoders (VAEs). The disentanglement of features in the latent space of VAEs allows us to discover feature-target correlations in datasets and semi-automatically evaluate them for ML shortcuts. We demonstrate the applicability of our method on several real-world datasets and identify shortcuts that have not been discovered before.