Home Research Papers Datasets State of the Art Pricing

Discover, visualize, and connect AI research papers. Explore the latest trends and insights in artificial intelligence research.

Product

Home
Research Papers
About

Support

Contact
Terms of Service
Privacy Policy

© 2026 Papersgraph. All rights reserved.

methodology-6

Meta-Learning

3260 papers • 126 benchmarks • 313 datasets

Meta-learning is a methodology considered with "learning to learn" machine learning algorithms. ( Image credit: Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks )

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in meta-learning-12

Trend

Dataset

Best Model

Actions

ML10

ML10

MT50

MT50

ML45

ML45

Libraries

i

Use these libraries to find meta-learning-12 models and implementations

learnables/learn2learn

12 papers 2,583

Datasets

Meta-Dataset

BlendedMVS

Meta-World Benchmark

Meta-World Benchmark

TyDiQA-GoldP

MQ2007

Subtasks

Few-Shot Learning Sample Probing universal meta-learning

Most implemented papers

Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks

Chelsea Finn, P. Abbeel, S. Levine•Wed Mar 08 2017

An algorithm for meta-learning that is model-agnostic, in the sense that it is compatible with any model trained with gradient descent and applicable to a variety of different learning problems, including classification, regression, and reinforcement learning is proposed.

13278

Content

Introduction Benchmarks Datasets Subtasks Libraries Papers

OMNIGLOT - 1-Shot, 20-way

OMNIGLOT - 1-Shot, 20-way

sicara/easy-few-shot-learning

8 papers 969

Guang000/Awesome-Dataset-Distillati…

7 papers 1,271

peymanbateni/simple-cnaps

6 papers 110

plai-group/simple-cnaps

6 papers 50

facebookresearch/higher

5 papers 1,579

cnguyen10/few_shot_meta_learning

5 papers 239

ExtremeWeather

ExtremeWeather

Kuzushiji-49

CODEBRIM

0

Prototypical Networks for Few-shot Learning

R. Zemel, Kevin Swersky, Jake Snell•Tue Mar 14 2017

This work proposes Prototypical Networks for few-shot classification, and provides an analysis showing that some simple design decisions can yield substantial improvements over recent approaches involving complicated architectural choices and meta-learning.

9429 0

Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples

H. Larochelle, Vincent Dumoulin, Pierre-Antoine Manzagol, Tyler Lixuan Zhu, Eleni Triantafillou, Pascal Lamblin, Kelvin Xu, Ross Goroshin, Carles Gelada, Kevin Swersky•Wed Mar 06 2019

This work proposes Meta-Dataset: a new benchmark for training and evaluating models that is large-scale, consists of diverse datasets, and presents more realistic tasks, and proposes a new set of baselines for quantifying the benefit of meta-learning in Meta- Dataset.

672 0

Learning to Compare: Relation Network for Few-Shot Learning

T. Xiang, Timothy M. Hospedales, Flood Sung, Yongxin Yang, Li Zhang, Philip H. S. Torr•Wed Nov 15 2017

A conceptually simple, flexible, and general framework for few-shot learning, where a classifier must learn to recognise new classes given only few examples from each, which is easily extended to zero- shot learning.

4413 0

On First-Order Meta-Learning Algorithms

John Schulman, Alex Nichol, Joshua Achiam•Wed Mar 07 2018

A family of algorithms for learning a parameter initialization that can be fine-tuned quickly on a new task, using only first-order derivatives for the meta-learning updates, including Reptile, which works by repeatedly sampling a task, training on it, and moving the initialization towards the trained weights on that task.

2478 0

A Closer Look at Few-shot Classification

Z. Kira, Wei-Yu Chen, Yen-Cheng Liu, Y. Wang, Jia-Bin Huang•Sun Apr 07 2019

The results reveal that reducing intra-class variation is an important factor when the feature backbone is shallow, but not as critical when using deeper backbones, and a baseline method with a standard fine-tuning practice compares favorably against other state-of-the-art few-shot learning algorithms.

1935 0

Meta-Baseline: Exploring Simple Meta-Learning for Few-Shot Learning

Trevor Darrell, Zhuang Liu, Xiaolong Wang, Yinbo Chen, Huijuan Xu•Sun Mar 08 2020

A simple process: meta-learning over a whole-classification pre-trained model on its evaluation metric achieves competitive performance to state-of-the-art methods on standard bench-marks and sheds some light on understanding the trade-offs between the meta- learning objective and the whole- classification objective in few-shot learning.

433 0

Learning to reinforcement learn

C. Blundell, M. Botvinick, D. Kumaran, Joel Z. Leibo, R. Munos, Hubert Soyer, Jane X. Wang, Z. Kurth-Nelson, Dhruva Tirumala•Wed Nov 16 2016

This work introduces a novel approach to deep meta-reinforcement learning, which is a system that is trained using one RL algorithm, but whose recurrent dynamics implement a second, quite separate RL procedure.

1037 0

Meta-Learning for Semi-Supervised Few-Shot Classification

H. Larochelle, R. Zemel, Mengye Ren, J. Tenenbaum, Eleni Triantafillou, Kevin Swersky, S. Ravi, Jake Snell•Wed Feb 14 2018

This work proposes novel extensions of Prototypical Networks that are augmented with the ability to use unlabeled examples when producing prototypes, and confirms that these models can learn to improve their predictions due to unlabeling examples, much like a semi-supervised algorithm would.

1399 0

Adding a benchmark result helps the community track progress.

Meta-Learning | State-of-the-Art