# Automated Relational Meta-learning

arXiv: https://arxiv.org/abs/2001.00745 · HTML: https://mlsanity.com/paper/2001.00745/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Jan 2020.

## Built on by (most cited first, 25 of 25)

- [A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities](https://mlsanity.com/paper/2205.06743/) (2022)
- [Meta-Knowledge Transfer for Inductive Knowledge Graph Embedding](https://mlsanity.com/paper/2110.14170/) (2021)
- [An Optimization-Based Meta-Learning Model for MRI Reconstruction with Diverse Dataset](https://mlsanity.com/paper/2110.00715/) (2021)
- [When Meta-Learning Meets Online and Continual Learning: A Survey](https://mlsanity.com/paper/2311.05241/) (2023)
- [Meta-Learning with a Geometry-Adaptive Preconditioner](https://mlsanity.com/paper/2304.01552/) (2023)
- [Knowledge-Aware Meta-learning for Low-Resource Text Classification](https://mlsanity.com/paper/2109.04707/) (2021)
- [Meta-learning in healthcare: A survey](https://mlsanity.com/paper/2308.02877/) (2023)
- [Conditional Hyper-Network for Blind Super-Resolution with Multiple Degradations](https://mlsanity.com/paper/2104.03926/) (2021)
- [Information Symmetry Matters: A Modal-Alternating Propagation Network for Few-Shot Learning](https://mlsanity.com/paper/2109.01295/) (2021)
- [Adaptive Feature Interpolation for Low-Shot Image Generation](https://mlsanity.com/paper/2112.02450/) (2021)
- [ST-MAML: A Stochastic-Task based Method for Task-Heterogeneous Meta-Learning](https://mlsanity.com/paper/2109.13305/) (2021)
- [Revisiting Few-Shot Learning from a Causal Perspective](https://mlsanity.com/paper/2209.13816/) (2022)
- [Semantic-Fused Multi-Granularity Cross-City Traffic Prediction](https://mlsanity.com/paper/2302.11774/) (2023)
- [Task-Distributionally Robust Data-Free Meta-Learning](https://mlsanity.com/paper/2311.14756/) (2023)
- [XB-MAML: Learning Expandable Basis Parameters for Effective Meta-Learning with Wide Task Coverage](https://mlsanity.com/paper/2403.06768/) (2024)
- [EMO: Episodic Memory Optimization for Few-Shot Meta-Learning](https://mlsanity.com/paper/2306.05189/) (2023)
- [Novel Representation Learning Technique using Graphs for Performance Analytics](https://mlsanity.com/paper/2401.10799/) (2024)
- [Generative Meta-Learning for Zero-Shot Relation Triplet Extraction](https://mlsanity.com/paper/2305.01920/) (2023)
- [Meta-learning Amidst Heterogeneity and Ambiguity](https://mlsanity.com/paper/2107.02228/) (2021)
- [Contrastive Knowledge-Augmented Meta-Learning for Few-Shot Classification](https://mlsanity.com/paper/2207.12346/) (2022)
- [Association Graph Learning for Multi-Task Classification with Category Shifts](https://mlsanity.com/paper/2210.04637/) (2022)
- [Learning to Generate Parameters of ConvNets for Unseen Image Data](https://mlsanity.com/paper/2310.11862/) (2023)
- [Exploring intra-task relations to improve meta-learning algorithms](https://mlsanity.com/paper/2312.16612/) (2023)
- [Doodle Your Keypoints: Sketch-Based Few-Shot Keypoint Detection](https://mlsanity.com/paper/2507.07994/) (2025)
- [Task-Level Contrastiveness for Cross-Domain Few-Shot Learning](https://mlsanity.com/paper/2510.03509/) (2025)

## Abstract (arXiv)

In order to efficiently learn with small amount of data on new tasks, meta-learning transfers knowledge learned from previous tasks to the new ones. However, a critical challenge in meta-learning is the task heterogeneity which cannot be well handled by traditional globally shared meta-learning methods. In addition, current task-specific meta-learning methods may either suffer from hand-crafted structure design or lack the capability to capture complex relations between tasks. In this paper, motivated by the way of knowledge organization in knowledge bases, we propose an automated relational meta-learning (ARML) framework that automatically extracts the cross-task relations and constructs the meta-knowledge graph. When a new task arrives, it can quickly find the most relevant structure and tailor the learned structure knowledge to the meta-learner. As a result, the proposed framework not only addresses the challenge of task heterogeneity by a learned meta-knowledge graph, but also increases the model interpretability. We conduct extensive experiments on 2D toy regression and few-shot image classification and the results demonstrate the superiority of ARML over state-of-the-art baselines.