Structural Adversarial Objectives for Self-Supervised Representation Learning
This addresses the need for self-supervised learning methods that avoid manual data augmentation, though it appears incremental as it builds on existing GAN frameworks.
The paper tackles the problem of self-supervised representation learning by proposing structural adversarial objectives within GANs, which guide the discriminator to extract informative features without relying on hand-crafted data augmentations, and experiments show it competes with contrastive learning methods on datasets like CIFAR-10/100 and an ImageNet subset.
Within the framework of generative adversarial networks (GANs), we propose objectives that task the discriminator for self-supervised representation learning via additional structural modeling responsibilities. In combination with an efficient smoothness regularizer imposed on the network, these objectives guide the discriminator to learn to extract informative representations, while maintaining a generator capable of sampling from the domain. Specifically, our objectives encourage the discriminator to structure features at two levels of granularity: aligning distribution characteristics, such as mean and variance, at coarse scales, and grouping features into local clusters at finer scales. Operating as a feature learner within the GAN framework frees our self-supervised system from the reliance on hand-crafted data augmentation schemes that are prevalent across contrastive representation learning methods. Across CIFAR-10/100 and an ImageNet subset, experiments demonstrate that equipping GANs with our self-supervised objectives suffices to produce discriminators which, evaluated in terms of representation learning, compete with networks trained by contrastive learning approaches.