# Why Do Animals Need Shaping? A Theory of Task Composition and Curriculum Learning

arXiv: https://arxiv.org/abs/2402.18361 · HTML: https://mlsanity.com/paper/2402.18361/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Feb 2024.

## Built on by (most cited first, 16 of 16)

- [A Survey of Agentic Reasoning for Large Language Models: Towards Recursively Self-Improving and Collective Agents](https://mlsanity.com/paper/2601.12538/) (2026)
- [Order parameters and phase transitions of continual learning in deep neural networks](https://mlsanity.com/paper/2407.10315/) (2024)
- [Optimal Protocols for Continual Learning via Statistical Physics and Control Theory](https://mlsanity.com/paper/2409.18061/) (2024)
- [Scaling can lead to compositional generalization](https://mlsanity.com/paper/2507.07207/) (2025)
- [Influence Dynamics and Stagewise Data Attribution](https://mlsanity.com/paper/2510.12071/) (2025)
- [A Theory of Initialisation's Impact on Specialisation](https://mlsanity.com/paper/2503.02526/) (2025)
- [Distinct Computations Emerge From Compositional Curricula in In-Context Learning](https://mlsanity.com/paper/2506.13253/) (2025)
- [Range, not Independence, Drives Modularity in Biologically Inspired Representations](https://mlsanity.com/paper/2410.06232/) (2024)
- [A statistical physics framework for optimal learning](https://mlsanity.com/paper/2507.07907/) (2025)
- [Flexible task abstractions emerge in linear networks with fast and bounded units](https://mlsanity.com/paper/2411.03840/) (2024)
- [Make Haste Slowly: A Theory of Emergent Structured Mixed Selectivity in Feature Learning ReLU Networks](https://mlsanity.com/paper/2503.06181/) (2025)
- [Compositional meta-learning through probabilistic task inference](https://mlsanity.com/paper/2510.01858/) (2025)
- [Learning to Solve Complex Problems via Dataset Decomposition](https://mlsanity.com/paper/2602.20296/) (2026)
- [Dual-Criterion Curriculum Learning: Application to Temporal Data](https://mlsanity.com/paper/2603.23573/) (2026)
- [AI, Trust, and Teaming: The Humans-as-Handlers Approach for Autonomous and Opaque AI Systems](https://mlsanity.com/paper/2607.00523/) (2026)
- [Pretraining Curricula Enable Selective Fine-tuning](https://mlsanity.com/paper/2607.04846/) (2026)

## Abstract (arXiv)

Diverse studies in systems neuroscience begin with extended periods of curriculum training known as `shaping' procedures. These involve progressively studying component parts of more complex tasks, and can make the difference between learning a task quickly, slowly or not at all. Despite the importance of shaping to the acquisition of complex tasks, there is as yet no theory that can help guide the design of shaping procedures, or more fundamentally, provide insight into its key role in learning. Modern deep reinforcement learning systems might implicitly learn compositional primitives within their multilayer policy networks. Inspired by these models, we propose and analyse a model of deep policy gradient learning of simple compositional reinforcement learning tasks. Using the tools of statistical physics, we solve for exact learning dynamics and characterise different learning strategies including primitives pre-training, in which task primitives are studied individually before learning compositional tasks. We find a complex interplay between task complexity and the efficacy of shaping strategies. Overall, our theory provides an analytical understanding of the benefits of shaping in a class of compositional tasks and a quantitative account of how training protocols can disclose useful task primitives, ultimately yielding faster and more robust learning.