paper-with-me

홈 › Papers

On Hard Episodes in Meta-Learning

2021-10-21 · Samyadeep Basu, Amr Sharaf, Nicolo Fusi, Soheil Feizi

Existing meta-learners primarily focus on improving the average task accuracy across multiple episodes. Different episodes, however, may vary in hardness and quality leading to a wide gap in the meta-learner's performance across episodes. Understanding this issue is particularly critical in industrial few-shot settings, where there is limited control over test episodes as they are typically uploaded by end-users. In this paper, we empirically analyse the behaviour of meta-learners on episodes of varying hardness across three standard benchmark datasets: CIFAR-FS, mini-ImageNet, and tiered-ImageNet. Surprisingly, we observe a wide gap in accuracy of around 50% between the hardest and easiest episodes across all the standard benchmarks and meta-learners. We additionally investigate various properties of hard episodes and highlight their connection to catastrophic forgetting during meta-training. To address the issue of sub-par performance on hard episodes, we investigate and benchmark different meta-training strategies based on adversarial training and curriculum learning. We find that adversarial training strategies are much more powerful than curriculum learning in improving the prediction performance on hard episodes.

📄 PDF Abstract BibTeX arXiv:2110.11190

Code (0)

등록된 구현이 없습니다.

Tasks

Meta-Learning

Methods 이 논문이 사용한 방법론

Test 설명 없음

Similar Papers 제목 키워드 기반

First-Explore, then Exploit: Meta-Learning to Solve Hard Exploration-Exploitation Trade-Offs

2023-07-05 · Ben Norman, Jeff Clune

Standard reinforcement learning (RL) agents never intelligently explore like a human (i.e. taking into account complex domain priors and adapting quickly based on previous exploration). Across episodes, RL agents struggl…

Meta-LearningReinforcement Learning (RL)

Unsupervised Few-Shot Action Recognition via Action-Appearance Aligned Meta-Adaptation

2021-09-30 · ICCV 2021 10 · Jay Patravali, Gaurav Mittal, Ye Yu, Fuxin Li 외

We present MetaUVFS as the first Unsupervised Meta-learning algorithm for Video Few-Shot action recognition. MetaUVFS leverages over 550K unlabeled videos to train a two-stream 2D and 3D CNN architecture via contrastive …

Action RecognitionContrastive LearningFew-Shot action recognitionFew Shot Action Recognition+1

Are Few-shot Learning Benchmarks Too Simple ?

2019-09-25 · Gabriel Huang, Hugo Larochelle, Simon Lacoste-Julien

We argue that the widely used Omniglot and miniImageNet benchmarks are too simple because their class semantics do not vary across episodes, which defeats their intended purpose of evaluating few-shot classification meth…

ClassificationFew-Shot Learning

A Hierarchical Bayesian Model for Deep Few-Shot Meta Learning

2023-06-16 · Minyoung Kim, Timothy Hospedales

We propose a novel hierarchical Bayesian model for learning with a large (possibly infinite) number of tasks/episodes, which suits well the few-shot meta learning problem. We consider episode-wise random variables to mod…

Bayesian InferenceMeta-LearningVariational Inference

Incremental Meta-Learning via Episodic Replay Distillation for Few-Shot Image Recognition

2021-11-09 · Kai Wang, Xialei Liu, Andy Bagdanov, Luis Herranz 외

Most meta-learning approaches assume the existence of a very large set of labeled data available for episodic meta-learning of base knowledge. This contrasts with the more realistic continual learning paradigm in which d…

Continual LearningKnowledge DistillationMeta-Learning