# Visual analogy: Deep learning versus compositional models

arXiv: https://arxiv.org/abs/2105.07065 · HTML: https://mlsanity.com/paper/2105.07065/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 29 references that are in the map.
- First posted on arXiv 14 May 2021.

## Builds on (most cited first, 13 of 13)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1802.02611/) (2018)
- [ShapeNet: An Information-Rich 3D Model Repository](https://mlsanity.com/paper/1512.03012/) (2015)
- [Identity Mappings in Deep Residual Networks](https://mlsanity.com/paper/1603.05027/) (2016)
- [Learning to Compare: Relation Network for Few-Shot Learning](https://mlsanity.com/paper/1711.06025/) (2017)
- [A simple neural network module for relational reasoning](https://mlsanity.com/paper/1706.01427/) (2017)
- [Detect What You Can: Detecting and Representing Objects using Holistic Models and Body Parts](https://mlsanity.com/paper/1406.2031/) (2014)
- [Measuring abstract reasoning in neural networks](https://mlsanity.com/paper/1807.04225/) (2018)
- [Learning Perceptual Inference by Contrasting](https://mlsanity.com/paper/1912.00086/) (2019)
- [Learning to Make Analogies by Contrasting Abstract Relational Structure](https://mlsanity.com/paper/1902.00120/) (2019)
- [VISALOGY: Answering Visual Analogy Questions](https://mlsanity.com/paper/1510.08973/) (2015)
- [Transforming task representations to perform novel tasks](https://mlsanity.com/paper/2005.04318/) (2020)

## Built on by (most cited first, 11 of 11)

- [Deep Learning Methods for Abstract Visual Reasoning: A Survey on Raven's Progressive Matrices](https://mlsanity.com/paper/2201.12382/) (2022)
- [Tell me why! Explanations support learning relational and causal structure](https://mlsanity.com/paper/2112.03753/) (2021)
- [A Review of Emerging Research Directions in Abstract Visual Reasoning](https://mlsanity.com/paper/2202.10284/) (2022)
- [Dissociating language and thought in large language models](https://mlsanity.com/paper/2301.06627/) (2023)
- [The Relational Bottleneck as an Inductive Bias for Efficient Abstraction](https://mlsanity.com/paper/2309.06629/) (2023)
- [Unfolding Spatial Cognition: Evaluating Multimodal Models on Visual Simulations](https://mlsanity.com/paper/2506.04633/) (2025)
- [Learning to Extrapolate: A Transductive Approach](https://mlsanity.com/paper/2304.14329/) (2023)
- [Reasoning Limitations of Multimodal Large Language Models. A Case Study of Bongard Problems](https://mlsanity.com/paper/2411.01173/) (2024)
- [Zero-shot visual reasoning through probabilistic analogical mapping](https://mlsanity.com/paper/2209.15087/) (2022)
- [Bongard-RWR+: Real-World Representations of Fine-Grained Concepts in Bongard Problems](https://mlsanity.com/paper/2508.12026/) (2025)
- [Imagine the Unseen World: A Benchmark for Systematic Generalization in Visual World Models](https://mlsanity.com/paper/2311.09064/) (2023)

## Abstract (arXiv)

Is analogical reasoning a task that must be learned to solve from scratch by applying deep learning models to massive numbers of reasoning problems? Or are analogies solved by computing similarities between structured representations of analogs? We address this question by comparing human performance on visual analogies created using images of familiar three-dimensional objects (cars and their subregions) with the performance of alternative computational models. Human reasoners achieved above-chance accuracy for all problem types, but made more errors in several conditions (e.g., when relevant subregions were occluded). We compared human performance to that of two recent deep learning models (Siamese Network and Relation Network) directly trained to solve these analogy problems, as well as to that of a compositional model that assesses relational similarity between part-based representations. The compositional model based on part representations, but not the deep learning models, generated qualitative performance similar to that of human reasoners.