Chroma-VAE: Mitigating Shortcut Learning with Generative Classifiers
This addresses shortcut learning, a critical issue in AI reliability, by offering a novel method to mitigate it, though it is incremental as it builds on existing generative models.
The paper tackles the problem of shortcut learning in deep neural networks by proposing Chroma-VAE, a generative classifier approach that isolates shortcuts in a small latent subspace, enabling training on shortcut-free features, resulting in improved performance on benchmark and real-world tasks.
Deep neural networks are susceptible to shortcut learning, using simple features to achieve low training loss without discovering essential semantic structure. Contrary to prior belief, we show that generative models alone are not sufficient to prevent shortcut learning, despite an incentive to recover a more comprehensive representation of the data than discriminative approaches. However, we observe that shortcuts are preferentially encoded with minimal information, a fact that generative models can exploit to mitigate shortcut learning. In particular, we propose Chroma-VAE, a two-pronged approach where a VAE classifier is initially trained to isolate the shortcut in a small latent subspace, allowing a secondary classifier to be trained on the complementary, shortcut-free latent subspace. In addition to demonstrating the efficacy of Chroma-VAE on benchmark and real-world shortcut learning tasks, our work highlights the potential for manipulating the latent space of generative classifiers to isolate or interpret specific correlations.