Barlow Twins: Self-Supervised Learning via Redundancy Reduction
Jure Žbontar, Jing Li, Ishan Misra et al.
2021 · arXiv (Cornell University) · 778 citations
Self-supervised learning (SSL) is rapidly closing the gap with supervised methods on large computer vision benchmarks. A successful approach to SSL is to learn embeddings which are invariant to distortions of the input sample. However, a recurring issue with this approach is the existence of trivial constant solutions. Most current methods avoid such solutions by careful implementation details. We propose an objective function that naturally avoids collapse by measuring the cross-correlation matrix between the outputs of two identical networks fed with distorted versions of a sample, and maki…
Explore this paper's citation graph on Constellation.