# Relational inductive bias for physical construction in humans and machines

arXiv: https://arxiv.org/abs/1806.01203 · HTML: https://mlsanity.com/paper/1806.01203/

## Key facts

- Cited by 44 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Jun 2018.

## Built on by (most cited first, 30 of 44)

- [Combinatorial optimization and reasoning with graph neural networks](https://mlsanity.com/paper/2102.09544/) (2021)
- [ETA Prediction with Graph Neural Networks in Google Maps](https://mlsanity.com/paper/2108.11482/) (2021)
- [The CLRS Algorithmic Reasoning Benchmark](https://mlsanity.com/paper/2205.15659/) (2022)
- [Beyond Pick-and-Place: Tackling Robotic Stacking of Diverse Shapes](https://mlsanity.com/paper/2110.06192/) (2021)
- [Learning Visible Connectivity Dynamics for Cloth Smoothing](https://mlsanity.com/paper/2105.10389/) (2021)
- [A Generalist Neural Algorithmic Learner](https://mlsanity.com/paper/2209.11142/) (2022)
- [CANOS: A Fast and Scalable Neural AC-OPF Solver Robust To N-1 Perturbations](https://mlsanity.com/paper/2403.17660/) (2024)
- [Dual Algorithmic Reasoning](https://mlsanity.com/paper/2302.04496/) (2023)
- [Brick-by-Brick: Combinatorial Construction with Deep Reinforcement Learning](https://mlsanity.com/paper/2110.15481/) (2021)
- [Neural Algorithmic Reasoning for Combinatorial Optimisation](https://mlsanity.com/paper/2306.06064/) (2023)
- [Large-scale graph representation learning with very deep GNNs and self-supervision](https://mlsanity.com/paper/2107.09422/) (2021)
- [Neural Relational Inference with Fast Modular Meta-learning](https://mlsanity.com/paper/2310.07015/) (2023)
- [Graph Reinforcement Learning for Combinatorial Optimization: A Survey and Unifying Perspective](https://mlsanity.com/paper/2404.06492/) (2024)
- [Persistent Message Passing](https://mlsanity.com/paper/2103.01043/) (2021)
- [Reasoning-Modulated Representations](https://mlsanity.com/paper/2107.08881/) (2021)
- [Asynchronous Algorithmic Alignment with Cocycles](https://mlsanity.com/paper/2306.15632/) (2023)
- [Algorithmic Concept-based Explainable Reasoning](https://mlsanity.com/paper/2107.07493/) (2021)
- [Active Learning of Abstract Plan Feasibility](https://mlsanity.com/paper/2107.00683/) (2021)
- [Constraining Generative Models for Engineering Design with Negative Data](https://mlsanity.com/paper/2306.15166/) (2023)
- [On the Markov Property of Neural Algorithmic Reasoning: Analyses and Methods](https://mlsanity.com/paper/2403.04929/) (2024)
- [Discrete Neural Algorithmic Reasoning](https://mlsanity.com/paper/2402.11628/) (2024)
- [Primal-Dual Neural Algorithmic Reasoning](https://mlsanity.com/paper/2505.24067/) (2025)
- [Neural Priority Queues for Graph Neural Networks](https://mlsanity.com/paper/2307.09660/) (2023)
- [Automated Cyber Defense with Generalizable Graph-based Reinforcement Learning Agents](https://mlsanity.com/paper/2509.16151/) (2025)
- [Deep Equilibrium Algorithmic Reasoning](https://mlsanity.com/paper/2410.15059/) (2024)
- [Feudal Graph Reinforcement Learning](https://mlsanity.com/paper/2304.05099/) (2023)
- [Reconstructing short-lived particles using hypergraph representation learning](https://mlsanity.com/paper/2402.10149/) (2024)
- [Compositional Law Parsing with Latent Random Functions](https://mlsanity.com/paper/2209.09115/) (2022)
- [Open-Book Neural Algorithmic Reasoning](https://mlsanity.com/paper/2501.00072/) (2024)
- [NAR-*ICP: Neural Execution of Classical ICP-based Pointcloud Registration Algorithms](https://mlsanity.com/paper/2410.11031/) (2024)

## Abstract (arXiv)

While current deep learning systems excel at tasks such as object classification, language processing, and gameplay, few can construct or modify a complex system such as a tower of blocks. We hypothesize that what these systems lack is a "relational inductive bias": a capacity for reasoning about inter-object relations and making choices over a structured description of a scene. To test this hypothesis, we focus on a task that involves gluing pairs of blocks together to stabilize a tower, and quantify how well humans perform. We then introduce a deep reinforcement learning agent which uses object- and relation-centric scene and policy representations and apply it to the task. Our results show that these structured representations allow the agent to outperform both humans and more naive approaches, suggesting that relational inductive bias is an important component in solving structured reasoning problems and for building more intelligent, flexible machines.