English

Testing using Privileged Information by Adapting Features with Statistical Dependence

Machine Learning 2021-11-05 v1 Computer Vision and Pattern Recognition

Abstract

Given an imperfect predictor, we exploit additional features at test time to improve the predictions made, without retraining and without knowledge of the prediction function. This scenario arises if training labels or data are proprietary, restricted, or no longer available, or if training itself is prohibitively expensive. We assume that the additional features are useful if they exhibit strong statistical dependence to the underlying perfect predictor. Then, we empirically estimate and strengthen the statistical dependence between the initial noisy predictor and the additional features via manifold denoising. As an example, we show that this approach leads to improvement in real-world visual attribute ranking. Project webpage: http://www.jamestompkin.com/tupi

Keywords

Cite

@article{arxiv.2111.02865,
  title  = {Testing using Privileged Information by Adapting Features with Statistical Dependence},
  author = {Kwang In Kim and James Tompkin},
  journal= {arXiv preprint arXiv:2111.02865},
  year   = {2021}
}

Comments

Published at ICCV 2021. Webpage: http://www.jamestompkin.com/tupi

R2 v1 2026-06-24T07:26:08.225Z