# Learning Task-General Representations with Generative Neuro-Symbolic Modeling

arXiv: https://arxiv.org/abs/2006.14448 · HTML: https://mlsanity.com/paper/2006.14448/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Jun 2020.

## Built on by (most cited first, 14 of 14)

- [Abstraction and Analogy-Making in Artificial Intelligence](https://mlsanity.com/paper/2102.10717/) (2021)
- [Neuro-Symbolic AI in 2024: A Systematic Review](https://mlsanity.com/paper/2501.05435/) (2025)
- [Unsupervised Learning of Neurosymbolic Encoders](https://mlsanity.com/paper/2107.13132/) (2021)
- [Compositional diversity in visual concept learning](https://mlsanity.com/paper/2305.19374/) (2023)
- [Neuro-Symbolic Learning: Principles and Applications in Ophthalmology](https://mlsanity.com/paper/2208.00374/) (2022)
- [Instilling Inductive Biases with Subnetworks](https://mlsanity.com/paper/2310.10899/) (2023)
- [Drawing out of Distribution with Neuro-Symbolic Generative Models](https://mlsanity.com/paper/2206.01829/) (2022)
- [Computing with Categories in Machine Learning](https://mlsanity.com/paper/2303.04156/) (2023)
- [Learning to Infer Generative Template Programs for Visual Concepts](https://mlsanity.com/paper/2403.15476/) (2024)
- [A Probabilistic Generative Model of Free Categories](https://mlsanity.com/paper/2205.04545/) (2022)
- [Character Generation through Self-Supervised Vectorization](https://mlsanity.com/paper/2208.02012/) (2022)
- [Certifying Global Robustness for Deep Neural Networks](https://mlsanity.com/paper/2405.20556/) (2024)
- [Hybrid Classical-Quantum architecture for vectorised image classification of hand-written sketches](https://mlsanity.com/paper/2407.06416/) (2024)
- [Bridging the Gap: Representation Spaces in Neuro-Symbolic AI](https://mlsanity.com/paper/2411.04393/) (2024)

## Abstract (arXiv)

People can learn rich, general-purpose conceptual representations from only raw perceptual inputs. Current machine learning approaches fall well short of these human standards, although different modeling traditions often have complementary strengths. Symbolic models can capture the compositional and causal knowledge that enables flexible generalization, but they struggle to learn from raw inputs, relying on strong abstractions and simplifying assumptions. Neural network models can learn directly from raw data, but they struggle to capture compositional and causal structure and typically must retrain to tackle new tasks. We bring together these two traditions to learn generative models of concepts that capture rich compositional and causal structure, while learning from raw data. We develop a generative neuro-symbolic (GNS) model of handwritten character concepts that uses the control flow of a probabilistic program, coupled with symbolic stroke primitives and a symbolic image renderer, to represent the causal and compositional processes by which characters are formed. The distributions of parts (strokes), and correlations between parts, are modeled with neural network subroutines, allowing the model to learn directly from raw data and express nonparametric statistical relationships. We apply our model to the Omniglot challenge of human-level concept learning, using a background set of alphabets to learn an expressive prior distribution over character drawings. In a subsequent evaluation, our GNS model uses probabilistic inference to learn rich conceptual representations from a single training image that generalize to 4 unique tasks, succeeding where previous work has fallen short.