Reproduce “Pairwise Learning Increases Data Efficiency in Deep Neural Networks” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Robert Bain, Marjan Rashidi, Elizabeth Chrastil

The authors introduce pairwise supervision for age estimation: a model learns which of two faces is older, then uses labeled references to estimate age. With 100 labeled faces, it achieves 13.9-year MAE versus 22.6 years for pointwise regression. A scratch-trained ResNet-10 reports 3.64-year MAE on 4,000 test faces; on STL-10, advantage is not significant at 100 labels.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.