Stable Diffusion Dataset Generation for Downstream Classification Tasks

2024/05/04 by Lomurno, Eugenio, D'Oria, Matteo, Matteucci, Matteo · 1 citation
#Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.2405.02698

Abstract

Recent advances in generative artificial intelligence have enabled the creation of high-quality synthetic data that closely mimics real-world data. This paper explores the adaptation of the Stable Diffusion 2.0 model for generating synthetic datasets, using Transfer Learning, Fine-Tuning and generation parameter optimisation techniques to improve the utility of the dataset for downstream classification tasks. We present a class-conditional version of the model that exploits a Class-Encoder and optimisation of key generation parameters. Our methodology led to synthetic datasets that, in a third of cases, produced models that outperformed those trained on real datasets.

Cited by

• On Approaches to Building Surrogate ODE Models for Diffusion Bridges

Related

• On the Limitation of Diffusion Models for Synthesizing Training Datasets
• Accelerating Diffusion Model Training under Minimal Budgets: A Condensation-Based Perspective
• Generative Modeling with Diffusion
• Learning Data Representations with Joint Diffusion Models
• Training on Thin Air: Improve Image Classification with Generated Data