Not All Unlabeled Data are Equal: Learning to Weight Data in Semi-supervised Learning
This work addresses the challenge of improving semi-supervised learning efficiency and accuracy for researchers and practitioners in machine learning, representing an incremental advancement over prior methods.
The paper tackles the problem of equally weighting unlabeled data in semi-supervised learning by proposing a method to assign different weights to each unlabeled example using an influence function-based algorithm, and it demonstrates that this technique outperforms state-of-the-art methods on image and language classification tasks.
Existing semi-supervised learning (SSL) algorithms use a single weight to balance the loss of labeled and unlabeled examples, i.e., all unlabeled examples are equally weighted. But not all unlabeled data are equal. In this paper we study how to use a different weight for every unlabeled example. Manual tuning of all those weights -- as done in prior work -- is no longer possible. Instead, we adjust those weights via an algorithm based on the influence function, a measure of a model's dependency on one training example. To make the approach efficient, we propose a fast and effective approximation of the influence function. We demonstrate that this technique outperforms state-of-the-art methods on semi-supervised image and language classification tasks.