Beyond Cropping and Rotation: Automated Evolution of Powerful Task-Specific Augmentations with Generative Models
This work addresses the problem of overfitting in vision models for researchers and practitioners by automating augmentation design, though it is incremental as it builds on existing generative models and AutoAugment methods.
The paper tackles the challenge of designing task-specific data augmentations that balance diversity and performance by introducing EvoAug, an automated pipeline that uses generative models and evolutionary algorithms to learn optimal augmentations, achieving strong results in fine-grained classification and few-shot learning tasks.
Data augmentation has long been a cornerstone for reducing overfitting in vision models, with methods like AutoAugment automating the design of task-specific augmentations. Recent advances in generative models, such as conditional diffusion and few-shot NeRFs, offer a new paradigm for data augmentation by synthesizing data with significantly greater diversity and realism. However, unlike traditional augmentations like cropping or rotation, these methods introduce substantial changes that enhance robustness but also risk degrading performance if the augmentations are poorly matched to the task. In this work, we present EvoAug, an automated augmentation learning pipeline, which leverages these generative models alongside an efficient evolutionary algorithm to learn optimal task-specific augmentations. Our pipeline introduces a novel approach to image augmentation that learns stochastic augmentation trees that hierarchically compose augmentations, enabling more structured and adaptive transformations. We demonstrate strong performance across fine-grained classification and few-shot learning tasks. Notably, our pipeline discovers augmentations that align with domain knowledge, even in low-data settings. These results highlight the potential of learned generative augmentations, unlocking new possibilities for robust model training.