# Learning Evolved Combinatorial Symbols with a Neuro-symbolic Generative Model

arXiv: https://arxiv.org/abs/2104.08274 · HTML: https://mlsanity.com/paper/2104.08274/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 18 references that are in the map.
- First posted on arXiv 16 Apr 2021.

## Builds on (most cited first, 4 of 4)

- [Sequence to Sequence Learning with Neural Networks](https://mlsanity.com/paper/1409.3215/) (2014)
- [Synthesizing Programs for Images using Reinforced Adversarial Learning](https://mlsanity.com/paper/1804.01118/) (2018)
- [Learning to learn generative programs with Memoised Wake-Sleep](https://mlsanity.com/paper/2007.03132/) (2020)
- [Generating new concepts with hybrid neuro-symbolic models](https://mlsanity.com/paper/2003.08978/) (2020)

## Built on by (most cited first, 3 of 3)

- [PoE-World: Compositional World Modeling with Products of Programmatic Experts](https://mlsanity.com/paper/2505.10819/) (2025)
- [A Survey on Compositional Learning of AI Models: Theoretical and Experimental Practices](https://mlsanity.com/paper/2406.08787/) (2024)
- [Sketching With Your Voice: "Non-Phonorealistic" Rendering of Sounds via Vocal Imitation](https://mlsanity.com/paper/2409.13507/) (2024)

## Abstract (arXiv)

Humans have the ability to rapidly understand rich combinatorial concepts from limited data. Here we investigate this ability in the context of auditory signals, which have been evolved in a cultural transmission experiment to study the emergence of combinatorial structure in language. We propose a neuro-symbolic generative model which combines the strengths of previous approaches to concept learning. Our model performs fast inference drawing on neural network methods, while still retaining the interpretability and generalization from limited data seen in structured generative approaches. This model outperforms a purely neural network-based approach on classification as evaluated against both ground truth and human experimental classification preferences, and produces superior reproductions of observed signals as well. Our results demonstrate the power of flexible combined neural-symbolic architectures for human-like generalization in raw perceptual domains and offers a step towards developing precise computational models of inductive biases in language evolution.