Adaptive Task Sampling for Meta-Learning

07/17/2020
by   Chenghao Liu, et al.
5

Meta-learning methods have been extensively studied and applied in computer vision, especially for few-shot classification tasks. The key idea of meta-learning for few-shot classification is to mimic the few-shot situations faced at test time by randomly sampling classes in meta-training data to construct few-shot tasks for episodic training. While a rich line of work focuses solely on how to extract meta-knowledge across tasks, we exploit the complementary problem on how to generate informative tasks. We argue that the randomly sampled tasks could be sub-optimal and uninformative (e.g., the task of classifying "dog" from "laptop" is often trivial) to the meta-learner. In this paper, we propose an adaptive task sampling method to improve the generalization performance. Unlike instance based sampling, task based sampling is much more challenging due to the implicit definition of the task in each episode. Therefore, we accordingly propose a greedy class-pair based sampling method, which selects difficult tasks according to class-pair potentials. We evaluate our adaptive task sampling method on two few-shot classification benchmarks, and it achieves consistent improvements across different feature backbones, meta-learning algorithms and datasets.

READ FULL TEXT
research
06/09/2022

Learning to generate imaginary tasks for improving generalization in meta-learning

The success of meta-learning on existing benchmarks is predicated on the...
research
01/27/2021

Similarity of Classification Tasks

Recent advances in meta-learning has led to remarkable performances on s...
research
02/11/2020

Incremental Learning for Metric-Based Meta-Learners

Majority of the modern meta-learning methods for few-shot classification...
research
06/04/2018

Meta Learner with Linear Nulling

We propose a meta learning algorithm utilizing a linear transformer that...
research
07/13/2020

Expert Training: Task Hardness Aware Meta-Learning for Few-Shot Classification

Deep neural networks are highly effective when a large number of labeled...
research
12/06/2021

Curriculum Meta-Learning for Few-shot Classification

We propose an adaptation of the curriculum training framework, applicabl...
research
03/23/2022

Multidimensional Belief Quantification for Label-Efficient Meta-Learning

Optimization-based meta-learning offers a promising direction for few-sh...

Please sign up or login with your details

Forgot password? Click here to reset