# A Meta Learning Approach to Discerning Causal Graph Structure

arXiv: https://arxiv.org/abs/2106.05859 · HTML: https://mlsanity.com/paper/2106.05859/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 14 references that are in the map.
- First posted on arXiv 6 Jun 2021.

## Builds on (most cited first, 9 of 9)

- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Deep Clustering for Unsupervised Learning of Visual Features](https://mlsanity.com/paper/1807.05520/) (2018)
- [Causal Effect Inference with Deep Latent-Variable Models](https://mlsanity.com/paper/1705.08821/) (2017)
- [A Meta-Transfer Objective for Learning to Disentangle Causal Mechanisms](https://mlsanity.com/paper/1901.10912/) (2019)
- [Online Deep Clustering for Unsupervised Representation Learning](https://mlsanity.com/paper/2006.10645/) (2020)
- [Causal Reasoning from Meta-reinforcement Learning](https://mlsanity.com/paper/1901.08162/) (2019)
- [Fairness Through Causal Awareness: Learning Latent-Variable Models for Biased Data](https://mlsanity.com/paper/1809.02519/) (2018)
- [Causal Generative Neural Networks](https://mlsanity.com/paper/1711.08936/) (2017)
- [Meta Learning for Causal Direction](https://mlsanity.com/paper/2007.02809/) (2020)

## Built on by (most cited first, 1 of 1)

- [Graph Attention Convolutional U-NET: A Semantic Segmentation Model for Identifying Flooded Areas](https://mlsanity.com/paper/2502.15907/) (2025)

## Abstract (arXiv)

We explore the usage of meta-learning to derive the causal direction between variables by optimizing over a measure of distribution simplicity. We incorporate a stochastic graph representation which includes latent variables and allows for more generalizability and graph structure expression. Our model is able to learn causal direction indicators for complex graph structures despite effects of latent confounders. Further, we explore robustness of our method with respect to violations of our distributional assumptions and data scarcity. Our model is particularly robust to modest data scarcity, but is less robust to distributional changes. By interpreting the model predictions as stochastic events, we propose a simple ensemble method classifier to reduce the outcome variability as an average of biased events. This methodology demonstrates ability to infer the existence as well as the direction of a causal relationship between data distributions.