# Transfer Learning with Label Noise

arXiv: https://arxiv.org/abs/1707.09724 · HTML: https://mlsanity.com/paper/1707.09724/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jul 2017.

## Built on by (most cited first, 4 of 4)

- [Estimating Instance-dependent Bayes-label Transition Matrix using a Deep Neural Network](https://mlsanity.com/paper/2105.13001/) (2021)
- [Improving Medical Image Classification with Label Noise Using Dual-uncertainty Estimation](https://mlsanity.com/paper/2103.00528/) (2021)
- [Joining datasets via data augmentation in the label space for neural networks](https://mlsanity.com/paper/2106.09260/) (2021)
- [Advancing Medical Image Segmentation: Morphology-Driven Learning with Diffusion Transformer](https://mlsanity.com/paper/2408.00347/) (2024)

## Abstract (arXiv)

Transfer learning aims to improve learning in target domain by borrowing knowledge from a related but different source domain. To reduce the distribution shift between source and target domains, recent methods have focused on exploring invariant representations that have similar distributions across domains. However, when learning this invariant knowledge, existing methods assume that the labels in source domain are uncontaminated, while in reality, we often have access to source data with noisy labels. In this paper, we first show how label noise adversely affect the learning of invariant representations and the correcting of label shift in various transfer learning scenarios. To reduce the adverse effects, we propose a novel Denoising Conditional Invariant Component (DCIC) framework, which provably ensures (1) extracting invariant representations given examples with noisy labels in source domain and unlabeled examples in target domain; (2) estimating the label distribution in target domain with no bias. Experimental results on both synthetic and real-world data verify the effectiveness of the proposed method.