Adversarial Auto-Augment with Label Preservation: A Representation Learning Principle Guided ApproachDownload PDF

Published: 31 Oct 2022, Last Modified: 28 May 2023NeurIPS 2022 AcceptReaders: Everyone
Keywords: Data Augmentation, Representation Learning, Adversarial Auto-Augment
TL;DR: We propose a label-preserving adversarial auto-augment method for data augmentation guided by a representation learning principle.
Abstract: Data augmentation is a critical contributing factor to the success of deep learning but heavily relies on prior domain knowledge which is not always available. Recent works on automatic data augmentation learn a policy to form a sequence of augmentation operations, which are still pre-defined and restricted to limited options. In this paper, we show that a prior-free autonomous data augmentation's objective can be derived from a representation learning principle that aims to preserve the minimum sufficient information of the labels. Given an example, the objective aims at creating a distant ``hard positive example'' as the augmentation, while still preserving the original label. We then propose a practical surrogate to the objective that can be optimized efficiently and integrated seamlessly into existing methods for a broad class of machine learning tasks, e.g., supervised, semi-supervised, and noisy-label learning. Unlike previous works, our method does not require training an extra generative model but instead leverages the intermediate layer representations of the end-task model for generating data augmentations. In experiments, we show that our method consistently brings non-trivial improvements to the three aforementioned learning tasks from both efficiency and final performance, either or not combined with pre-defined augmentations, e.g., on medical images when domain knowledge is unavailable and the existing augmentation techniques perform poorly. Code will be released publicly.
Supplementary Material: pdf
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 1 code implementation](
16 Replies