When to Learn What: Model-Adaptive Data Augmentation Curriculum

• URL: http://arxiv.org/abs/2309.04747v2
• Date: Sat, 30 Sep 2023 06:14:36 GMT
• Title: When to Learn What: Model-Adaptive Data Augmentation Curriculum
• Authors: Chengkai Hou, Jieyu Zhang, Tianyi Zhou
• Abstract summary: We propose Model Adaptive Data Augmentation (MADAug) to jointly train an augmentation policy network to teach the model when to learn what. Unlike previous work, MADAug selects augmentation operators for each input image by a model-adaptive policy varying between training stages, producing a data augmentation curriculum optimized for better generalization.
• Score: 32.99634881669643
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Data augmentation (DA) is widely used to improve the generalization of neural networks by enforcing the invariances and symmetries to pre-defined transformations applied to input data. However, a fixed augmentation policy may have different effects on each sample in different training stages but existing approaches cannot adjust the policy to be adaptive to each sample and the training model. In this paper, we propose Model Adaptive Data Augmentation (MADAug) that jointly trains an augmentation policy network to teach the model when to learn what. Unlike previous work, MADAug selects augmentation operators for each input image by a model-adaptive policy varying between training stages, producing a data augmentation curriculum optimized for better generalization. In MADAug, we train the policy through a bi-level optimization scheme, which aims to minimize a validation-set loss of a model trained using the policy-produced data augmentations. We conduct an extensive evaluation of MADAug on multiple image classification tasks and network architectures with thorough comparisons to existing DA approaches. MADAug outperforms or is on par with other baselines and exhibits better fairness: it brings improvement to all classes and more to the difficult ones. Moreover, MADAug learned policy shows better performance when transferred to fine-grained datasets. In addition, the auto-optimized policy in MADAug gradually introduces increasing perturbations and naturally forms an easy-to-hard curriculum.

Related papers

• AdaAugment: A Tuning-Free and Adaptive Approach to Enhance Data Augmentation [12.697608744311122]
  AdaAugment is a tuning-free Adaptive Augmentation method. It dynamically adjusts augmentation magnitudes for individual training samples based on real-time feedback from the target network. It consistently outperforms other state-of-the-art DA methods in effectiveness while maintaining remarkable efficiency.
  arXiv  Detail & Related papers  (2024-05-19T06:54:03Z)
• Genetic Learning for Designing Sim-to-Real Data Augmentations [1.03590082373586]
  Data augmentations are useful in closing the sim-to-real domain gap when training on synthetic data. Many image augmentation techniques exist, parametrized by different settings, such as strength and probability. This paper presents two different interpretable metrics that can be combined to predict how well a certain augmentation policy will work for a specific sim-to-real setting.
  arXiv  Detail & Related papers  (2024-03-11T15:00:56Z)
• Learning to Augment via Implicit Differentiation for Domain Generalization [107.9666735637355]
  Domain generalization (DG) aims to overcome the problem by leveraging multiple source domains to learn a domain-generalizable model. In this paper, we propose a novel augmentation-based DG approach, dubbed AugLearn. AugLearn shows effectiveness on three standard DG benchmarks, PACS, Office-Home and Digits-DG.
  arXiv  Detail & Related papers  (2022-10-25T18:51:51Z)
• Augmentation Learning for Semi-Supervised Classification [13.519613713213277]
  We propose a Semi-Supervised Learning method that automatically selects the most effective data augmentation policy for a particular dataset. We show how policy learning can be used to adapt augmentations to datasets beyond ImageNet.
  arXiv  Detail & Related papers  (2022-08-03T10:06:51Z)
• Universal Adaptive Data Augmentation [30.83891617679216]
  "Universal Adaptive Data Augmentation" (UADA) is a novel data augmentation strategy. We randomly decide types and magnitudes of DA operations for every data batch during training. UADA adaptively update DA's parameters according to the target model's gradient information.
  arXiv  Detail & Related papers  (2022-07-14T05:05:43Z)
• Latent-Variable Advantage-Weighted Policy Optimization for Offline RL [70.01851346635637]
  offline reinforcement learning methods hold the promise of learning policies from pre-collected datasets without the need to query the environment for new transitions. In practice, offline datasets are often heterogeneous, i.e., collected in a variety of scenarios. We propose to leverage latent-variable policies that can represent a broader class of policy distributions. Our method improves the average performance of the next best-performing offline reinforcement learning methods by 49% on heterogeneous datasets.
  arXiv  Detail & Related papers  (2022-03-16T21:17:03Z)
• CADDA: Class-wise Automatic Differentiable Data Augmentation for EEG Signals [92.60744099084157]
  We propose differentiable data augmentation amenable to gradient-based learning. We demonstrate the relevance of our approach on the clinically relevant sleep staging classification task.
  arXiv  Detail & Related papers  (2021-06-25T15:28:48Z)
• ParticleAugment: Sampling-Based Data Augmentation [80.44268663372233]
  We propose a particle filtering formulation to find optimal augmentation policies and their schedules during model training. We show that our formulation for automated augmentation reaches promising results on CIFAR-10, CIFAR-100, and ImageNet datasets.
  arXiv  Detail & Related papers  (2021-06-16T10:56:02Z)
• Policy Learning with Adaptively Collected Data [22.839095992238537]
  We address the challenge of learning the optimal policy with adaptively collected data. We propose an algorithm based on generalized augmented inverse propensity weighted estimators. We demonstrate our algorithm's effectiveness using both synthetic data and public benchmark datasets.
  arXiv  Detail & Related papers  (2021-05-05T22:03:10Z)
• Learning Representational Invariances for Data-Efficient Action Recognition [52.23716087656834]
  We show that our data augmentation strategy leads to promising performance on the Kinetics-100, UCF-101, and HMDB-51 datasets. We also validate our data augmentation strategy in the fully supervised setting and demonstrate improved performance.
  arXiv  Detail & Related papers  (2021-03-30T17:59:49Z)
• CoDA: Contrast-enhanced and Diversity-promoting Data Augmentation for Natural Language Understanding [67.61357003974153]
  We propose a novel data augmentation framework dubbed CoDA. CoDA synthesizes diverse and informative augmented examples by integrating multiple transformations organically. A contrastive regularization objective is introduced to capture the global relationship among all the data samples.
  arXiv  Detail & Related papers  (2020-10-16T23:57:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.