mlsanityData to 14 Oct 2026

Fighting Selection Bias in Statistical Learning: Application to Visual Recognition from Biased Image Databases

arXiv 2109.02357 ↗First posted 6 Sep 2021cs.CV cs.CY cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

In practice, and especially when training deep neural networks, visual recognition rules are often learned based on various sources of information. On the other hand, the recent deployment of facial recognition systems with uneven performances on different population segments has highlighted the representativeness issues induced by a naive aggregation of the datasets. In this paper, we show how biasing models can remedy these problems. Based on the (approximate) knowledge of the biasing mechanisms at work, our approach consists in reweighting the observations, so as to form a nearly debiased estimator of the target distribution. One key condition is that the supports of the biased distributions must partly overlap, and cover the support of the target distribution. In order to meet this requirement in practice, we propose to use a low dimensional image representation, shared across the image databases. Finally, we provide numerical experiments highlighting the relevance of our approach.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 20 of its 41 references that are in the map. First posted on arXiv 6 Sep 2021.

Builds on

20 of 41 references.

References · 20

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2015Domain-Adversarial Training of Neural Networkscited by 3,044
  4. 2016Matching Networks for One Shot Learningcited by 2,255
  5. 2015Learning Transferable Features with Deep Adaptation Networkscited by 1,329
  6. 2017Learning to Generalize: Meta-Learning for Domain Generalizationcited by 763
  7. 2016MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognitioncited by 624
  8. 2020Long-tail learning via logit adjustmentcited by 618
  9. 2016Unsupervised Domain Adaptation with Residual Transfer Networkscited by 341
  10. 2016Domain Separation Networkscited by 320
  11. 2018Learning Not to Learn: Training Deep Neural Networks with Biased Datacited by 255
  12. 2018Women also Snowboard: Overcoming Bias in Captioning Modelscited by 236
8 more
  1. 2015Simultaneous Deep Transfer Across Domains and Taskscited by 206
  2. 2018Large Scale Fine-Grained Categorization and Domain-Specific Transfer Learningcited by 174
  3. 2018Efficient parametrization of multi-domain deep neural networkscited by 168
  4. 2018SpotTune: Transfer Learning through Adaptive Fine-tuningcited by 168
  5. 2017Borrowing Treasures from the Wealthy: Deep Transfer Learning through Selective Joint Fine-tuningcited by 48
  6. 2018Domain Adaptive Transfer Learning with Specialist Modelscited by 31
  7. 2020Weighted Empirical Risk Minimization: Sample Selection Bias Correction based on Importance Samplingcited by 7
  8. 2019Statistical Learning from Biased Training Samplescited by 4

Built on by

Not cited yet.