FROST: Faster and more Robust One-shot Semi-supervised Training

TL;DR

FROST introduces a faster, more robust one-shot semi-supervised learning method that outperforms existing approaches in speed and resilience to data and hyper-parameter variations.

Contribution

The paper presents FROST, a novel one-shot semi-supervised training approach combining self-training with a single network, enhancing speed and robustness over prior methods.

Findings

FROST trains up to ten times faster than state-of-the-art methods.

FROST maintains high performance even with unknown unlabeled data composition.

FROST is less sensitive to hyper-parameter choices.

Abstract

Recent advances in one-shot semi-supervised learning have lowered the barrier for deep learning of new applications. However, the state-of-the-art for semi-supervised learning is slow to train and the performance is sensitive to the choices of the labeled data and hyper-parameter values. In this paper, we present a one-shot semi-supervised learning method that trains up to an order of magnitude faster and is more robust than state-of-the-art methods. Specifically, we show that by combining semi-supervised learning with a one-stage, single network version of self-training, our FROST methodology trains faster and is more robust to choices for the labeled samples and changes in hyper-parameters. Our experiments demonstrate FROST's capability to perform well when the composition of the unlabeled data is unknown; that is when the unlabeled data contain unequal numbers of each class and can contain out-of-distribution examples that don't belong to any of the training classes. High performance, speed of training, and insensitivity to hyper-parameters make FROST the most practical method for one-shot semi-supervised training. Our code is available at https://github.com/HelenaELiu/FROST.