Recent progress has shown that few-shot learning can be improved with access to unlabelled data, known as semi-supervised few-shot learning(SS-FSL). We introduce an SS-FSL approach, dubbed as Prototypical Random Walk Networks(PRWN), built on top of Prototypical Networks (PN). We develop a random walk semi-supervised loss that enables the network to learn representations that are compact and well-separated. Our work is related to the very recent development on graph-based approaches for few-shot learning. However, we show that compact and well-separated class representations can be achieved by modeling our prototypical random walk notion without needing additional graph-NN parameters or requiring a transductive setting where collective test set is provided. Our model outperforms prior art in most benchmarks with significant improvements in some cases. For example, in a mini-Imagenet 5-shot classification task, we obtain 69.65% accuracy to the 64.59% state-of-the-art. Our model, trained with 40% of the data as labelled, compares competitively against fully supervised prototypical networks, trained on 100% of the labels, even outperforming it in the 1-shot mini-Imagenet case with 50.89% to 49.4% accuracy. We also show that our model is resistant to distractors, unlabeled data that does not belong to any of the training classes, and hence reflecting robustness to labelled/unlabelled class distribution mismatch. We also performed a challenging discriminative power test, showing a relative improvement on top of the baseline of ≈14% on 20 classes on mini-Imagenet and ≈60% on 800 classes on Omniglot. Code will be made available.
Interactive explainer
Auto-generated. This explainer was generated automatically with Claude (Anthropic) from the published paper. It simplifies the method and may contain errors — for a correct understanding, and for all results and numbers, please refer to the published paper (publisher version).
@inproceedings{ayyad2021semi,
author = {Ayyad, Ahmed and Li, Yuchen and Muaz, Raden and Albarqouni, Shadi and Elhoseiny, Mohamed},
booktitle = {AAAI Workshop on Meta Learning},
title = {Semi-Supervised Few-Shot Learning with Prototypical Random Walks},
year = {2021}
}
Privacy — With your consent we use Google Analytics to understand how our website is used (IP addresses anonymised).
You can change your choice at any time via “Cookie settings” in the footer. Privacy policy