Reinforcement Learning-Guided Semi-Supervised Learning

TL;DR

This paper introduces RLGSSL, a reinforcement learning-guided semi-supervised learning method that adaptively leverages labeled and unlabeled data, outperforming existing SSL techniques through a novel reward-based approach.

Contribution

It proposes a reinforcement learning framework for SSL, formulating it as a bandit problem with a new reward function, and employs a teacher-student model for improved stability.

Findings

RLGSSL achieves superior performance on benchmark datasets.

The RL-guided approach effectively balances labeled and unlabeled data.

Experimental results outperform state-of-the-art SSL methods.

Abstract

In recent years, semi-supervised learning (SSL) has gained significant attention due to its ability to leverage both labeled and unlabeled data to improve model performance, especially when labeled data is scarce. However, most current SSL methods rely on heuristics or predefined rules for generating pseudo-labels and leveraging unlabeled data. They are limited to exploiting loss functions and regularization methods within the standard norm. In this paper, we propose a novel Reinforcement Learning (RL) Guided SSL method, RLGSSL, that formulates SSL as a one-armed bandit problem and deploys an innovative RL loss based on weighted reward to adaptively guide the learning process of the prediction model. RLGSSL incorporates a carefully designed reward function that balances the use of labeled and unlabeled data to enhance generalization performance. A semi-supervised teacher-student framework is further deployed to increase the learning stability. We demonstrate the effectiveness of RLGSSL through extensive experiments on several benchmark datasets and show that our approach achieves consistent superior performance compared to state-of-the-art SSL methods.