# Gradient-Based Meta-Learning with Learned Layerwise Metric and Subspace

arXiv: https://arxiv.org/abs/1801.05558 · HTML: https://mlsanity.com/paper/1801.05558/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Jan 2018.

## Built on by (most cited first, 4 of 4)

- [Investigating Bi-Level Optimization for Learning and Vision from a Unified Perspective: A Survey and Beyond](https://mlsanity.com/paper/2101.11517/) (2021)
- [Towards Gradient-based Bilevel Optimization with Non-convex Followers and Beyond](https://mlsanity.com/paper/2110.00455/) (2021)
- [Incremental Few-Shot Learning via Implanting and Compressing](https://mlsanity.com/paper/2203.10297/) (2022)
- [Task-Specific Preconditioner for Cross-Domain Few-Shot Learning](https://mlsanity.com/paper/2412.15483/) (2024)

## Abstract (arXiv)

Gradient-based meta-learning methods leverage gradient descent to learn the commonalities among various tasks. While previous such methods have been successful in meta-learning tasks, they resort to simple gradient descent during meta-testing. Our primary contribution is the em MT-net, which enables the meta-learner to learn on each layer's activation space a subspace that the task-specific learner performs gradient descent on. Additionally, a task-specific learner of an em MT-net performs gradient descent with respect to a meta-learned distance metric, which warps the activation space to be more sensitive to task identity. We demonstrate that the dimension of this learned subspace reflects the complexity of the task-specific learner's adaptation task, and also that our model is less sensitive to the choice of initial learning rates than previous gradient-based meta-learning methods. Our method achieves state-of-the-art or comparable performance on few-shot classification and regression tasks.