# Improve Learning from Crowds via Generative Augmentation

arXiv: https://arxiv.org/abs/2107.10449 · HTML: https://mlsanity.com/paper/2107.10449/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 41 references that are in the map.
- First posted on arXiv 22 Jul 2021.

## Builds on (most cited first, 18 of 18)

- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets](https://mlsanity.com/paper/1606.03657/) (2016)
- [Data Augmentation Generative Adversarial Networks](https://mlsanity.com/paper/1711.04340/) (2017)
- [Human uncertainty makes classification more robust](https://mlsanity.com/paper/1908.07086/) (2019)
- [Learning From Noisy Labels By Regularized Estimation Of Annotator Confusion](https://mlsanity.com/paper/1902.03680/) (2019)
- [GraphGAN: Graph Representation Learning with Generative Adversarial Nets](https://mlsanity.com/paper/1711.08267/) (2017)
- [Deep learning from crowds](https://mlsanity.com/paper/1709.01779/) (2017)
- [Semi-Supervised Learning with Generative Adversarial Networks](https://mlsanity.com/paper/1606.01583/) (2016)
- [Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networks](https://mlsanity.com/paper/1511.06390/) (2015)
- [Who Said What: Modeling Individual Labelers Improves Classification](https://mlsanity.com/paper/1703.08774/) (2017)
- [IRGAN: A Minimax Game for Unifying Generative and Discriminative Information Retrieval Models](https://mlsanity.com/paper/1705.10513/) (2017)
- [Improving Generalization by Controlling Label-Noise Information in Neural Network Weights](https://mlsanity.com/paper/2002.07933/) (2020)
- [Near-Optimally Teaching the Crowd to Classify](https://mlsanity.com/paper/1402.2092/) (2014)
- [Learning from Crowds by Modeling Common Confusions](https://mlsanity.com/paper/2012.13052/) (2020)
- [Max-MIG: an Information Theoretic Approach for Joint Learning from Crowds](https://mlsanity.com/paper/1905.13436/) (2019)
- [Neural Message Passing for Multi-Label Classification](https://mlsanity.com/paper/1904.08049/) (2019)
- [Analysis of Minimax Error Rate for Crowdsourcing and Its Application to Worker Clustering Model](https://mlsanity.com/paper/1802.04551/) (2018)
- [Leveraging GPT-2 for Classifying Spam Reviews with Limited Labeled Data via Adversarial Training](https://mlsanity.com/paper/2012.13400/) (2020)

## Built on by (most cited first, 6 of 6)

- [Meta Policy Learning for Cold-Start Conversational Recommendation](https://mlsanity.com/paper/2205.11788/) (2022)
- [Robust Deep Learning from Crowds with Belief Propagation](https://mlsanity.com/paper/2111.00734/) (2021)
- [Mixture of Experts based Multi-task Supervise Learning from Crowds](https://mlsanity.com/paper/2407.13268/) (2024)
- [Improving a Named Entity Recognizer Trained on Noisy Data with a Few Clean Instances](https://mlsanity.com/paper/2310.16790/) (2023)
- [Understanding the Essence: Delving into Annotator Prototype Learning for Multi-Class Annotation Aggregation](https://mlsanity.com/paper/2508.02123/) (2025)
- [AHEAD: Advancing Multi-Class Label Aggregation with Interpretable Cross-Annotator Modeling](https://mlsanity.com/paper/2607.18465/) (2026)

## Abstract (arXiv)

Crowdsourcing provides an efficient label collection schema for supervised machine learning. However, to control annotation cost, each instance in the crowdsourced data is typically annotated by a small number of annotators. This creates a sparsity issue and limits the quality of machine learning models trained on such data. In this paper, we study how to handle sparsity in crowdsourced data using data augmentation. Specifically, we propose to directly learn a classifier by augmenting the raw sparse annotations. We implement two principles of high-quality augmentation using Generative Adversarial Networks: 1) the generated annotations should follow the distribution of authentic ones, which is measured by a discriminator; 2) the generated annotations should have high mutual information with the ground-truth labels, which is measured by an auxiliary network. Extensive experiments and comparisons against an array of state-of-the-art learning from crowds methods on three real-world datasets proved the effectiveness of our data augmentation framework. It shows the potential of our algorithm for low-budget crowdsourcing in general.