🤖 AI Summary
Addressing semi-supervised image classification under label scarcity, this paper proposes Meta Co-Training: a framework that leverages pretrained models to efficiently construct complementary dual-view representations—eliminating reliance on handcrafted, conditionally independent views required by classical co-training. It integrates meta-learning to dynamically select high-confidence pseudo-labels and performs collaborative optimization of the two models via meta-gradient updates, thereby mitigating pseudo-label noise accumulation. This work is the first to extend the Meta Pseudo Labels paradigm to dual-view co-training, offering both theoretical rigor and computational efficiency. Evaluated on ImageNet-10%, it achieves new state-of-the-art performance with minimal training resources. Moreover, it significantly outperforms existing semi-supervised methods across multiple fine-grained classification benchmarks.
📝 Abstract
In many practical computer vision scenarios unlabeled data is plentiful, but labels are scarce and difficult to obtain. As a result, semi-supervised learning which leverages unlabeled data to boost the performance of supervised classifiers have received significant attention in recent literature. One major class of semi-supervised algorithms is co-training. In co-training two different models leverage different independent and sufficient"views"of the data to jointly make better predictions. During co-training each model creates pseudo labels on unlabeled points which are used to improve the other model. We show that in the common case when independent views are not available we can construct such views inexpensively using pre-trained models. Co-training on the constructed views yields a performance improvement over any of the individual views we construct and performance comparable with recent approaches in semi-supervised learning, but has some undesirable properties. To alleviate the issues present with co-training we present Meta Co-Training which is an extension of the successful Meta Pseudo Labels approach to two views. Our method achieves new state-of-the-art performance on ImageNet-10% with very few training resources, as well as outperforming prior semi-supervised work on several other fine-grained image classification datasets.