How Sensitive are Meta-Learners to Dataset Imbalance?
This addresses the problem of dataset imbalance in meta-learning for researchers, showing an implicit strength but is incremental as it builds on known issues in supervised methods.
The study investigated how meta-learning methods for few-shot learning are affected by imbalanced meta-datasets, finding that they are more robust to meta-dataset imbalance than task-level imbalance, with effects persisting even under high imbalance ratios up to 65.
Meta-Learning (ML) has proven to be a useful tool for training Few-Shot Learning (FSL) algorithms by exposure to batches of tasks sampled from a meta-dataset. However, the standard training procedure overlooks the dynamic nature of the real-world where object classes are likely to occur at different frequencies. While it is generally understood that imbalanced tasks harm the performance of supervised methods, there is no significant research examining the impact of imbalanced meta-datasets on the FSL evaluation task. This study exposes the magnitude and extent of this problem. Our results show that ML methods are more robust against meta-dataset imbalance than imbalance at the task-level with a similar imbalance ratio ($ρ<20$), with the effect holding even in long-tail datasets under a larger imbalance ($ρ=65$). Overall, these results highlight an implicit strength of ML algorithms, capable of learning generalizable features under dataset imbalance and domain-shift. The code to reproduce the experiments is released under an open-source license.