# Modular meta-learning

arXiv: https://arxiv.org/abs/1806.10166 · HTML: https://mlsanity.com/paper/1806.10166/

## Key facts

- Cited by 51 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Jun 2018.

## Built on by (most cited first, 30 of 51)

- [The CLRS Algorithmic Reasoning Benchmark](https://mlsanity.com/paper/2205.15659/) (2022)
- [Modular Deep Learning](https://mlsanity.com/paper/2302.11529/) (2023)
- [MetaAvatar: Learning Animatable Clothed Human Models from Few Depth Images](https://mlsanity.com/paper/2106.11944/) (2021)
- [Wide Neural Networks Forget Less Catastrophically](https://mlsanity.com/paper/2110.11526/) (2021)
- [Combining Induction and Transduction for Abstract Reasoning](https://mlsanity.com/paper/2411.02272/) (2024)
- [Modular Lifelong Reinforcement Learning via Neural Composition](https://mlsanity.com/paper/2207.00429/) (2022)
- [Meta-Learning Dynamics Forecasting Using Task Inference](https://mlsanity.com/paper/2102.10271/) (2021)
- [Fast and Slow Learning of Recurrent Independent Mechanisms](https://mlsanity.com/paper/2105.08710/) (2021)
- [Continual Learning with Recursive Gradient Optimization](https://mlsanity.com/paper/2201.12522/) (2022)
- [Combining Modular Skills in Multitask Learning](https://mlsanity.com/paper/2202.13914/) (2022)
- [How to Reuse and Compose Knowledge for a Lifetime of Tasks: A Survey on Continual Learning and Functional Composition](https://mlsanity.com/paper/2207.07730/) (2022)
- [Understanding the Dynamics of DNNs Using Graph Modularity](https://mlsanity.com/paper/2111.12485/) (2021)
- [Deep Multi-Task Learning for Joint Localization, Perception, and Prediction](https://mlsanity.com/paper/2101.06720/) (2021)
- [CompoSuite: A Compositional Reinforcement Learning Benchmark](https://mlsanity.com/paper/2207.04136/) (2022)
- [Learning Connectivity with Graph Convolutional Networks for Skeleton-based Action Recognition](https://mlsanity.com/paper/2112.03328/) (2021)
- [Transformers with Competitive Ensembles of Independent Mechanisms](https://mlsanity.com/paper/2103.00336/) (2021)
- [On the link between conscious function and general intelligence in humans and machines](https://mlsanity.com/paper/2204.05133/) (2022)
- [AdaptNet: Policy Adaptation for Physics-Based Character Control](https://mlsanity.com/paper/2310.00239/) (2023)
- [Neural Relational Inference with Fast Modular Meta-learning](https://mlsanity.com/paper/2310.07015/) (2023)
- [Modular Meta-Learning for Power Control via Random Edge Graph Neural Networks](https://mlsanity.com/paper/2108.13178/) (2021)
- [Relay Hindsight Experience Replay: Self-Guided Continual Reinforcement Learning for Sequential Object Manipulation Tasks with Sparse Rewards](https://mlsanity.com/paper/2208.00843/) (2022)
- [Beyond Data Filtering: Knowledge Localization for Capability Removal in LLMs](https://mlsanity.com/paper/2512.05648/) (2025)
- [Quantifying Local Specialization in Deep Neural Networks](https://mlsanity.com/paper/2110.08058/) (2021)
- [Mnemosyne: Learning to Train Transformers with Transformers](https://mlsanity.com/paper/2302.01128/) (2023)
- [Mixture-of-Variational-Experts for Continual Learning](https://mlsanity.com/paper/2110.12667/) (2021)
- [Compositional Learning-based Planning for Vision POMDPs](https://mlsanity.com/paper/2112.09456/) (2021)
- [On the Generalization and Adaption Performance of Causal Models](https://mlsanity.com/paper/2206.04620/) (2022)
- [Physics of Skill Learning](https://mlsanity.com/paper/2501.12391/) (2025)
- [Transfering Hierarchical Structure with Dual Meta Imitation Learning](https://mlsanity.com/paper/2201.11981/) (2022)
- [Egocentric Planning for Scalable Embodied Task Achievement](https://mlsanity.com/paper/2306.01295/) (2023)

## Abstract (arXiv)

Many prediction problems, such as those that arise in the context of robotics, have a simplifying underlying structure that, if known, could accelerate learning. In this paper, we present a strategy for learning a set of neural network modules that can be combined in different ways. We train different modular structures on a set of related tasks and generalize to new tasks by composing the learned modules in new ways. By reusing modules to generalize we achieve combinatorial generalization, akin to the "infinite use of finite means" displayed in language. Finally, we show this improves performance in two robotics-related problems.