LGMLOct 3, 2019

Generalized Inner Loop Meta-Learning

arXiv:1910.01727v2167 citations
Originality Incremental advance
AI Analysis

This work provides a foundational framework for meta-learning researchers, though it is incremental in formalizing existing patterns.

The paper formalizes a common pattern in meta-learning called GIMLI, proves its requirements, and derives a general-purpose algorithm, along with releasing a library called higher to support future research.

Many (but not all) approaches self-qualifying as "meta-learning" in deep learning and reinforcement learning fit a common pattern of approximating the solution to a nested optimization problem. In this paper, we give a formalization of this shared pattern, which we call GIMLI, prove its general requirements, and derive a general-purpose algorithm for implementing similar approaches. Based on this analysis and algorithm, we describe a library of our design, higher, which we share with the community to assist and enable future research into these kinds of meta-learning approaches. We end the paper by showcasing the practical applications of this framework and library through illustrative experiments and ablation studies which they facilitate.

Code Implementations3 repos
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes