# Challenges in Disentangling Independent Factors of Variation

arXiv: https://arxiv.org/abs/1711.02245 · HTML: https://mlsanity.com/paper/1711.02245/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Nov 2017.

## Built on by (most cited first, 10 of 10)

- [ImageBART: Bidirectional Context with Multinomial Diffusion for Autoregressive Image Synthesis](https://mlsanity.com/paper/2108.08827/) (2021)
- [Scaling-up Disentanglement for Image Translation](https://mlsanity.com/paper/2103.14017/) (2021)
- [A Contrastive Objective for Learning Disentangled Representations](https://mlsanity.com/paper/2203.11284/) (2022)
- [Rethinking Content and Style: Exploring Bias for Unsupervised Disentanglement](https://mlsanity.com/paper/2102.10544/) (2021)
- [Retriever: Learning Content-Style Representation as a Token-Level Bipartite Graph](https://mlsanity.com/paper/2202.12307/) (2022)
- [Causally Disentangled Generative Variational AutoEncoder](https://mlsanity.com/paper/2302.11737/) (2023)
- [FaIRCoP: Facial Image Retrieval using Contrastive Personalization](https://mlsanity.com/paper/2205.15870/) (2022)
- [ProtoVAE: Prototypical Networks for Unsupervised Disentanglement](https://mlsanity.com/paper/2305.09092/) (2023)
- [Disentangling stellar atmospheric parameters in astronomical spectra using Generative Adversarial Neural Networks](https://mlsanity.com/paper/2501.11762/) (2025)
- [RobustSurg: Tackling domain generalisation for out-of-distribution surgical scene segmentation](https://mlsanity.com/paper/2512.02188/) (2025)

## Abstract (arXiv)

We study the problem of building models that disentangle independent factors of variation. Such models could be used to encode features that can efficiently be used for classification and to transfer attributes between different images in image synthesis. As data we use a weakly labeled training set. Our weak labels indicate what single factor has changed between two data samples, although the relative value of the change is unknown. This labeling is of particular interest as it may be readily available without annotation costs. To make use of weak labels we introduce an autoencoder model and train it through constraints on image pairs and triplets. We formally prove that without additional knowledge there is no guarantee that two images with the same factor of variation will be mapped to the same feature. We call this issue the reference ambiguity. Moreover, we show the role of the feature dimensionality and adversarial training. We demonstrate experimentally that the proposed model can successfully transfer attributes on several datasets, but show also cases when the reference ambiguity occurs.