Prompt-Driven Feature Diffusion for Open-World Semi-Supervised Learning
This addresses the problem of semi-supervised learning in open-world scenarios where new classes appear without labels, representing a novel method for a known bottleneck.
The paper tackles the challenge of lacking labeled data for unseen classes in Open World Semi-Supervised Learning by introducing Prompt-Driven Feature Diffusion (PDFD), which uses class-specific prompts in a diffusion model to generate discriminative features, resulting in remarkable performance enhancements over state-of-the-art methods.
In this paper, we present a novel approach termed Prompt-Driven Feature Diffusion (PDFD) within a semi-supervised learning framework for Open World Semi-Supervised Learning (OW-SSL). At its core, PDFD deploys an efficient feature-level diffusion model with the guidance of class-specific prompts to support discriminative feature representation learning and feature generation, tackling the challenge of the non-availability of labeled data for unseen classes in OW-SSL. In particular, PDFD utilizes class prototypes as prompts in the diffusion model, leveraging their class-discriminative and semantic generalization ability to condition and guide the diffusion process across all the seen and unseen classes. Furthermore, PDFD incorporates a class-conditional adversarial loss for diffusion model training, ensuring that the features generated via the diffusion process can be discriminatively aligned with the class-conditional features of the real data. Additionally, the class prototypes of the unseen classes are computed using only unlabeled instances with confident predictions within a semi-supervised learning framework. We conduct extensive experiments to evaluate the proposed PDFD. The empirical results show PDFD exhibits remarkable performance enhancements over many state-of-the-art existing methods.