# Hierarchical Relational Inference

arXiv: https://arxiv.org/abs/2010.03635 · HTML: https://mlsanity.com/paper/2010.03635/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Oct 2020.

## Built on by (most cited first, 9 of 9)

- [VAEL: Bridging Variational Autoencoders and Probabilistic Logic Programming](https://mlsanity.com/paper/2202.04178/) (2022)
- [Systematic Visual Reasoning through Object-Centric Relational Abstraction](https://mlsanity.com/paper/2306.02500/) (2023)
- [Learning to Generalize with Object-centric Agents in the Open World Survival Game Crafter](https://mlsanity.com/paper/2208.03374/) (2022)
- [Recurrent Complex-Weighted Autoencoders for Unsupervised Object Discovery](https://mlsanity.com/paper/2405.17283/) (2024)
- [Unsupervised Image Decomposition with Phase-Correlation Networks](https://mlsanity.com/paper/2110.03473/) (2021)
- [Learning Tree Structures from Leaves For Particle Decay Reconstruction](https://mlsanity.com/paper/2208.14924/) (2022)
- [Boxhead: A Dataset for Learning Hierarchical Representations](https://mlsanity.com/paper/2110.03628/) (2021)
- [Leveraging Hidden Structure in Self-Supervised Learning](https://mlsanity.com/paper/2106.16060/) (2021)
- [VideoPCDNet: Video Parsing and Prediction with Phase Correlation Networks](https://mlsanity.com/paper/2506.19621/) (2025)

## Abstract (arXiv)

Common-sense physical reasoning in the real world requires learning about the interactions of objects and their dynamics. The notion of an abstract object, however, encompasses a wide variety of physical objects that differ greatly in terms of the complex behaviors they support. To address this, we propose a novel approach to physical reasoning that models objects as hierarchies of parts that may locally behave separately, but also act more globally as a single whole. Unlike prior approaches, our method learns in an unsupervised fashion directly from raw visual images to discover objects, parts, and their relations. It explicitly distinguishes multiple levels of abstraction and improves over a strong baseline at modeling synthetic and real-world videos.