Shot Learning Benchmark

Shot learning benchmarks evaluate algorithms' ability to learn new tasks from limited data, a crucial area for advancing artificial intelligence. Current research focuses on comparing meta-learning approaches, like Model-Agnostic Meta-Learning (MAML), with transfer learning methods, investigating the impact of dataset diversity and exploring architectural improvements such as specialized LSTMs. These benchmarks are vital for objectively assessing progress in few-shot learning, informing the development of more efficient and robust AI systems applicable to various domains where data scarcity is a limiting factor.

Papers

October 22, 2023

Are LSTMs Good Few-Shot Learners?
Mike Huisman, Thomas M. Moerland, Aske Plaat, Jan N. van Rijn
LeArning Abstract Meta Learning LSTM Network Shot Learning Benchmark

June 24, 2023

Is Pre-training Truly Better Than Meta-Learning?
Brando Miranda, Patrick Yu, Saumya Goyal, Yu-Xiong Wang, Sanmi Koyejo
Meta Learning Shot Learning Benchmark Shot Meta Learning Meta Training Stage

June 14, 2023

Improving Generalization in Meta-Learning via Meta-Gradient Augmentation
Ren Wang, Haoliang Sun, Qi Wei, Xiushan Nie, Yuling Ma, Yilong Yin
Strong Generalization Meta Learning Gradient Regularization Meta Gradient Memorization Capacity Shot Learning Benchmark

March 2, 2023

STUNT: Few-shot Tabular Learning with Self-generated Tasks from Unlabeled Tables
Jaehyun Nam, Jihoon Tack, Kyungmin Lee, Hankook Lee, Jinwoo Shin
Shot Learning Tabular Datasets Tabular Learning Task Model Shot Learning Benchmark Self Generated Shot Tabular

December 1, 2022

Finetune like you pretrain: Improved finetuning of zero-shot vision models
Sachin Goyal, Ananya Kumar, Sankalp Garg, Zico Kolter, Aditi Raghunathan
Zero Shot Fine Tuning Contrastive Loss Image Text Model Supervised Finetuning Contrastive Fine Tuning Shot Learning Benchmark

October 10, 2022

Multi-Modal Fusion by Meta-Initialization
Matthew T. Jackson, Shreshth A. Malik, Michael T. Matthews, Yousuf Mohamed-Ahmed
Multi Modal Fusion Task Inference Shot Learning Benchmark Meta Initialization Driven Initialization Agnostic Meta Learning

August 26, 2022

Expanding continual few-shot learning benchmarks to include recognition of specific instances
Gideon Kowadlo, Abdelrahman Ahmed, Amir Mayan, David Rawlinson
Machine Learning Continual LEArning Shot Learning Recognition Rate Shot Learning Benchmark

August 2, 2022

The Curse of Low Task Diversity: On the Failure of Transfer Learning to Outperform MAML and Their Empirical Equivalence
Brando Miranda, Patrick Yu, Yu-Xiong Wang, Sanmi Koyejo
Transfer Learning Reversal Curse Agnostic Meta learnIng Model Equivalence Task Diversity Shot Learning Benchmark Meta Learning Benchmark

April 26, 2022

Meta-free few-shot learning via representation learning with weight averaging
Kuilin Chen, Chi-Guhn Lee
Transfer Learning Shot Learning Individual Representation Shot Classification Meta Level WEight AVERaging Shot Learning Benchmark

March 17, 2022

Global Convergence of MAML and Theory-Inspired Neural Architecture Search for Few-Shot Learning
Haoxiang Wang, Yite Wang, Ruoyu Sun, Bo Li
Neural Network LeArning Abstract Shot Learning Neural Architecture Search Global Convergence Model Agnostic Meta Learning Shot Learning Benchmark

January 18, 2022

Instance-aware Prompt Learning for Language Understanding and Generation
Feihu Jin, Jinliang Lu, Jiajun Zhang, Chengqing Zong
Pre Trained Language Model Language Understanding Faithful Generation Prompt Learning Prompt Token Shot Learning Benchmark

December 24, 2021