1S-DAug: One-Shot Data Augmentation for Robust Few-Shot Generalization
This addresses the challenge of robust generalization in few-shot learning for AI systems, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the problem of few-shot learning by introducing 1S-DAug, a one-shot generative augmentation operator that synthesizes diverse image variants from a single example at test time, achieving up to 20% proportional accuracy improvement on the miniImagenet benchmark.
Few-shot learning (FSL) challenges model generalization to novel classes based on just a few shots of labeled examples, a testbed where traditional test-time augmentations fail to be effective. We introduce 1S-DAug, a one-shot generative augmentation operator that synthesizes diverse yet faithful variants from just one example image at test time. 1S-DAug couples traditional geometric perturbations with controlled noise injection and a denoising diffusion process conditioned on the original image. The generated images are then encoded and aggregated, alongside the original image, into a combined representation for more robust FSL predictions. Integrated as a training-free model-agnostic plugin, 1S-DAug consistently improves FSL across standard benchmarks of 4 different datasets without any model parameter update, including achieving up to 20\% proportional accuracy improvement on the miniImagenet 5-way-1-shot benchmark. Code will be released.