Improvements to context based self-supervised learning

TL;DR

This paper introduces new methods to enhance context-based self-supervised learning, addressing issues like chromatic aberration and spatial skew, resulting in top benchmark scores across various datasets and tasks.

Contribution

The authors develop novel techniques to improve self-supervised learning performance and generalization, achieving state-of-the-art results on multiple benchmarks.

Findings

Top scores on all standard self-supervised benchmarks

4.0 to 7.1 percentage point improvement over baseline

Demonstrated generalization across different architectures

Abstract

We develop a set of methods to improve on the results of self-supervised learning using context. We start with a baseline of patch based arrangement context learning and go from there. Our methods address some overt problems such as chromatic aberration as well as other potential problems such as spatial skew and mid-level feature neglect. We prevent problems with testing generalization on common self-supervised benchmark tests by using different datasets during our development. The results of our methods combined yield top scores on all standard self-supervised benchmarks, including classification and detection on PASCAL VOC 2007, segmentation on PASCAL VOC 2012, and "linear tests" on the ImageNet and CSAIL Places datasets. We obtain an improvement over our baseline method of between 4.0 to 7.1 percentage points on transfer learning classification tests. We also show results on different standard network architectures to demonstrate generalization as well as portability. All data, models and programs are available at: https://gdo-datasci.llnl.gov/selfsupervised/.