Universal Representation Learning from Multiple Domains for Few-shot Classification
This work addresses the problem of few-shot classification for AI systems needing to adapt quickly to new domains with limited data, representing an incremental advance over prior methods.
The paper tackles few-shot classification across unseen classes and domains by learning a single set of universal deep representations through knowledge distillation and co-alignment, achieving significant performance improvements and greater efficiency on the Meta-Dataset benchmark.
In this paper, we look at the problem of few-shot classification that aims to learn a classifier for previously unseen classes and domains from few labeled samples. Recent methods use adaptation networks for aligning their features to new domains or select the relevant features from multiple domain-specific feature extractors. In this work, we propose to learn a single set of universal deep representations by distilling knowledge of multiple separately trained networks after co-aligning their features with the help of adapters and centered kernel alignment. We show that the universal representations can be further refined for previously unseen domains by an efficient adaptation step in a similar spirit to distance learning methods. We rigorously evaluate our model in the recent Meta-Dataset benchmark and demonstrate that it significantly outperforms the previous methods while being more efficient. Our code will be available at https://github.com/VICO-UoE/URL.