Semi-supervised transfer learning with hierarchical self-regularization

Xingjian Li

doi:10.1016/j.patcog.2023.109831

Semi-supervised transfer learning with hierarchical self-regularization

2023-12-01

7

PRE

AI

DOI：10.1016/j.patcog.2023.109831

原文链接

原文求助

摘要

En 中文

Both semi-supervised learning and transfer learning aim at lowering the annotation burden for training models. However, such two tasks are usually studied separately, i.e. most semi-supervised learning algorithms train models from scratch while transfer learning assumes pre-trained models as the initialization. In this work, we focus on a previously-less-concerned setting that further reduces the annotation efforts through incorporating both semi-supervised and transfer learning, where specifically a pre-trained source model is used as the initialization of semi-supervised learning. As those powerful pre-trained models are ubiquitously available nowadays and can considerably benefit various down-streaming tasks, such a setting is relevant to real-world applications yet challenging to design effective algorithms.Aiming at enabling transfer learning under semi-supervised settings, we propose a hierarchical self regularization mechanism to exploit unlabeled samples more effectively, where a novel self-regularizer has been introduced to incorporate both individual-level and population-level regularization terms. The former term employs self-distillation to regularize learned deep features for each individual sample, and the latter one enforces self-consistency on feature distributions between labeled and unlabeled samples. Samples involved in both regularizers are weighted by an adaptive strategy, where self-regularization effects of both terms are adaptively controlled by the confidence of every sample. To validate our algorithm, exhaustive experiments have been conducted on diverse datasets such as CIFAR-10 for general object recognition, CUB200-2011/MIT-indoor-67 for fine-grained classification and MURA for medical image classification. Compared with state-of-the-art semi-supervised learning methods including Pseudo Label, Mean Teacher, MixMatch and FixMatch, our algorithm demonstrates two advantages: first of all, the proposed approach adopts a new point of view to tackle problems caused by inadequate supervision and achieves very competitive results; then, it is complementary to these state-of-the-art methods and thus can be combined with them to get additional improvements. Furthermore, our method can also be applied to fully supervised transfer learning and self supervised learning. We have published our code at https://github.com/SHI-Labs/Semi-Supervised-TransferLearning.

Keyword:

Semi-supervised learning

Transfer learning

Fine-tuning

Deep learning

Hierarchical consistency

Adaptive sample selection

Semi-supervised transfer learning with hierarchical self-regularization

摘要

期刊

机构