Shuffle to Learn: Self-supervised learning from permutations via differentiable ranking

Self-supervised pre-training using so-called "pretext" tasks has recently shown impressive performance across a wide range of tasks. In this work we advance self-supervised learning from permutations, that consists in shuffling parts of input and training a model to reorder them, improving downstream performance in classification. To do so, we overcome the main challenges of integrating permutation inversions (a discontinuous operation) into an end-to-end training scheme, heretofore sidestepped by casting the reordering task as classification, fundamentally reducing the space of permutations that can be exploited. These advances rely on two main, independent contributions. First, we use recent advances in differentiable ranking to integrate the permutation inversion flawlessly into a neural network, enabling us to use the full set of permutations, at no additional computing cost. Our experiments validate that learning from all possible permutations (up to $10^{18}$) improves the quality of the pre-trained representations over using a limited, fixed set. Second, we successfully demonstrate that inverting permutations is a meaningful pretext task in a diverse range of modalities, beyond images, which does not require modality-specific design. In particular, we also improve music understanding by reordering spectrogram patches in the frequency space, as well as video classification by reordering frames along the time axis. We furthermore analyze the influence of the patches that we use (vertical, horizontal, 2-dimensional), as well as the benefit of our approach in different data regimes.