AASAE: Augmentation-Augmented Stochastic Autoencoders
This work addresses self-supervised learning for image classification, offering a novel method that is largely comparable to state-of-the-art algorithms on some benchmarks, though it appears incremental as an alternative to existing paradigms.
The paper tackled self-supervised learning by proposing augmentation-augmented stochastic autoencoders (AASAE), which replace KL divergence regularization in VAEs with data augmentations to encode domain-specific invariances, resulting in performance improvements of 30% on CIFAR-10, 40% on STL-10, and 45% on Imagenet compared to VAEs.
Recent methods for self-supervised learning can be grouped into two paradigms: contrastive and non-contrastive approaches. Their success can largely be attributed to data augmentation pipelines which generate multiple views of a single input that preserve the underlying semantics. In this work, we introduce augmentation-augmented stochastic autoencoders (AASAE), yet another alternative to self-supervised learning, based on autoencoding. We derive AASAE starting from the conventional variational autoencoder (VAE), by replacing the KL divergence regularization, which is agnostic to the input domain, with data augmentations that explicitly encourage the internal representations to encode domain-specific invariances and equivariances. We empirically evaluate the proposed AASAE on image classification, similar to how recent contrastive and non-contrastive learning algorithms have been evaluated. Our experiments confirm the effectiveness of data augmentation as a replacement for KL divergence regularization. The AASAE outperforms the VAE by 30% on CIFAR-10, 40% on STL-10 and 45% on Imagenet. On CIFAR-10 and STL-10, the results for AASAE are largely comparable to the state-of-the-art algorithms for self-supervised learning.