Uniform Sampling over Episode Difficulty
arXiv:2108.01662
Abstract
Episodic training is a core ingredient of few-shot learning to train models on tasks with limited labelled data. Despite its success, episodic training remains largely understudied, prompting us to ask the question: what is the best way to sample episodes? In this paper, we first propose a method to approximate episode sampling distributions based on their difficulty. Building on this method, we perform an extensive analysis and find that sampling uniformly over episode difficulty outperforms other sampling schemes, including curriculum and easy-/hard-mining. As the proposed sampling method is algorithm agnostic, we can leverage these insights to improve few-shot learning accuracies across many episodic training algorithms. We demonstrate the efficacy of our method across popular few-shot learning datasets, algorithms, network architectures, and protocols.
arXiv commentNeurIPS'21 camera ready
References in corpus (20)
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Prototypical Networks for Few-shot Learning
- Prioritized Experience Replay
- Theoretical Models of Learning to Learn
- On First-Order Meta-Learning Algorithms
- DropBlock: A regularization method for convolutional networks
- Meta-Learning with Implicit Gradients
- TADAM: Task dependent adaptive metric for improved few-shot learning
- Recasting Gradient-Based Meta-Learning as Hierarchical Bayes
- Meta-learning with differentiable closed-form solvers
- Not All Samples Are Created Equal: Deep Learning with Importance Sampling
- Accelerating Minibatch Stochastic Gradient Descent using Stratified Sampling
- Meta-Learning with Warped Gradient Descent
- learn2learn: A Library for Meta-Learning Research
- Revisiting Meta-Learning as Supervised Learning
- On Episodes, Prototypical Networks, and Few-shot Learning
- Expert Training: Task Hardness Aware Meta-Learning for Few-Shot Classification
- When Do Curricula Work?
- Adaptive Task Sampling for Meta-Learning