# Max-MIG: an Information Theoretic Approach for Joint Learning from Crowds

arXiv: https://arxiv.org/abs/1905.13436 · HTML: https://mlsanity.com/paper/1905.13436/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 May 2019.

## Built on by (most cited first, 26 of 26)

- [End-to-End Weak Supervision](https://mlsanity.com/paper/2107.02233/) (2021)
- [Benchmarking LLMs' Judgments with No Gold Standard](https://mlsanity.com/paper/2411.07127/) (2024)
- [Trustable Co-label Learning from Multiple Noisy Annotators](https://mlsanity.com/paper/2203.04199/) (2022)
- [Deep Learning From Crowdsourced Labels: Coupled Cross-entropy Minimization, Identifiability, and Regularization](https://mlsanity.com/paper/2306.03288/) (2023)
- [Knowledge Learning with Crowdsourcing: A Brief Review and Systematic Perspective](https://mlsanity.com/paper/2206.09315/) (2022)
- [Transferring Annotator- and Instance-dependent Transition Matrix for Learning from Crowds](https://mlsanity.com/paper/2306.03116/) (2023)
- [Coupled Confusion Correction: Learning from Crowds with Sparse Annotations](https://mlsanity.com/paper/2312.07331/) (2023)
- [Diversified and Personalized Multi-rater Medical Image Segmentation](https://mlsanity.com/paper/2403.13417/) (2024)
- [Learning self-calibrated optic disc and cup segmentation from multi-rater annotations](https://mlsanity.com/paper/2206.05092/) (2022)
- [Multi-annotator Deep Learning: A Probabilistic Framework for Classification](https://mlsanity.com/paper/2304.02539/) (2023)
- [Improve Learning from Crowds via Generative Augmentation](https://mlsanity.com/paper/2107.10449/) (2021)
- [Learning from Noisy Labels via Conditional Distributionally Robust Optimization](https://mlsanity.com/paper/2411.17113/) (2024)
- [Learning to Complement with Multiple Humans](https://mlsanity.com/paper/2311.13172/) (2023)
- [Learning from Multiple Annotators by Incorporating Instance Features](https://mlsanity.com/paper/2106.15146/) (2021)
- [Opinions Vary? Diagnosis First!](https://mlsanity.com/paper/2202.06505/) (2022)
- [Calibrate the inter-observer segmentation uncertainty via diagnosis-first principle](https://mlsanity.com/paper/2208.03016/) (2022)
- [Neural-Hidden-CRF: A Robust Weakly-Supervised Sequence Labeler](https://mlsanity.com/paper/2309.05086/) (2023)
- [Learning From Crowdsourced Noisy Labels: A Signal Processing Perspective](https://mlsanity.com/paper/2407.06902/) (2024)
- [Multi-rater Prism: Learning self-calibrated medical image segmentation from multiple raters](https://mlsanity.com/paper/2212.00601/) (2022)
- [Learning from Noisy Crowd Labels with Logics](https://mlsanity.com/paper/2302.06337/) (2023)
- [A Unified Evaluation Framework for Multi-Annotator Tendency Learning](https://mlsanity.com/paper/2508.10393/) (2025)
- [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)
- [Probabilistic Label Spreading: Efficient and Consistent Estimation of Soft Labels with Epistemic Uncertainty on Graphs](https://mlsanity.com/paper/2602.04574/) (2026)
- [crowd-hpo: Realistic Hyperparameter Optimization and Benchmarking for Learning from Crowds with Noisy Labels](https://mlsanity.com/paper/2504.09085/) (2025)
- [Unsupervised Ensemble Learning Through Deep Energy-based Models](https://mlsanity.com/paper/2601.20556/) (2026)

## Abstract (arXiv)

Eliciting labels from crowds is a potential way to obtain large labeled data. Despite a variety of methods developed for learning from crowds, a key challenge remains unsolved: learning from crowds without knowing the information structure among the crowds a priori, when some people of the crowds make highly correlated mistakes and some of them label effortlessly (e.g. randomly). We propose an information theoretic approach, Max-MIG, for joint learning from crowds, with a common assumption: the crowdsourced labels and the data are independent conditioning on the ground truth. Max-MIG simultaneously aggregates the crowdsourced labels and learns an accurate data classifier. Furthermore, we devise an accurate data-crowds forecaster that employs both the data and the crowdsourced labels to forecast the ground truth. To the best of our knowledge, this is the first algorithm that solves the aforementioned challenge of learning from crowds. In addition to the theoretical validation, we also empirically show that our algorithm achieves the new state-of-the-art results in most settings, including the real-world data, and is the first algorithm that is robust to various information structures. Codes are available at hyperlinkhttps://github.com/Newbeeer/Max-MIGhttps://github.com/Newbeeer/Max-MIG