When Neural Networks Fail to Generalize? A Model Sensitivity Perspective

• URL: http://arxiv.org/abs/2212.00850v1
• Date: Thu, 1 Dec 2022 20:15:15 GMT
• Title: When Neural Networks Fail to Generalize? A Model Sensitivity Perspective
• Authors: Jiajin Zhang, Hanqing Chao, Amit Dhurandhar, Pin-Yu Chen, Ali Tajer, Yangyang Xu, Pingkun Yan
• Abstract summary: Domain generalization (DG) aims to train a model to perform well in unseen domains under different distributions. This paper considers a more realistic yet more challenging scenario, namely Single Domain Generalization (Single-DG) We empirically ascertain a property of a model that correlates strongly with its generalization that we coin as "model sensitivity" We propose a novel strategy of Spectral Adversarial Data Augmentation (SADA) to generate augmented images targeted at the highly sensitive frequencies.
• Score: 82.36758565781153
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Domain generalization (DG) aims to train a model to perform well in unseen domains under different distributions. This paper considers a more realistic yet more challenging scenario,namely Single Domain Generalization (Single-DG), where only a single source domain is available for training. To tackle this challenge, we first try to understand when neural networks fail to generalize? We empirically ascertain a property of a model that correlates strongly with its generalization that we coin as "model sensitivity". Based on our analysis, we propose a novel strategy of Spectral Adversarial Data Augmentation (SADA) to generate augmented images targeted at the highly sensitive frequencies. Models trained with these hard-to-learn samples can effectively suppress the sensitivity in the frequency space, which leads to improved generalization performance. Extensive experiments on multiple public datasets demonstrate the superiority of our approach, which surpasses the state-of-the-art single-DG methods.

Related papers

• Activate and Reject: Towards Safe Domain Generalization under Category Shift [71.95548187205736]
  We study a practical problem of Domain Generalization under Category Shift (DGCS) It aims to simultaneously detect unknown-class samples and classify known-class samples in the target domains. Compared to prior DG works, we face two new challenges: 1) how to learn the concept of unknown'' during training with only source known-class samples, and 2) how to adapt the source-trained model to unseen environments.
  arXiv  Detail & Related papers  (2023-10-07T07:53:12Z)
• Spectral Adversarial MixUp for Few-Shot Unsupervised Domain Adaptation [72.70876977882882]
  Domain shift is a common problem in clinical applications, where the training images (source domain) and the test images (target domain) are under different distributions. We propose a novel method for Few-Shot Unsupervised Domain Adaptation (FSUDA), where only a limited number of unlabeled target domain samples are available for training.
  arXiv  Detail & Related papers  (2023-09-03T16:02:01Z)
• CNN Feature Map Augmentation for Single-Source Domain Generalization [6.053629733936548]
  Domain Generalization (DG) has gained significant traction during the past few years. The goal in DG is to produce models which continue to perform well when presented with data distributions different from the ones available during training. We propose an alternative regularization technique for convolutional neural network architectures in the single-source DG image classification setting.
  arXiv  Detail & Related papers  (2023-05-26T08:48:17Z)
• Augmentation-based Domain Generalization for Semantic Segmentation [2.179313476241343]
  Unsupervised Domain Adaptation (UDA) and domain generalization (DG) aim to tackle the lack of generalization of Deep Neural Networks (DNNs) towards unseen domains. We study the in- and out-of-domain generalization capabilities of simple, rule-based image augmentations like blur, noise, color jitter and many more. Our experiments confirm the common scientific standard that combination of multiple different augmentations out-performs single augmentations.
  arXiv  Detail & Related papers  (2023-04-24T14:26:53Z)
• Domain Generalization with Adversarial Intensity Attack for Medical Image Segmentation [27.49427483473792]
  In real-world scenarios, it is common for models to encounter data from new and different domains to which they were not exposed to during training. domain generalization (DG) is a promising direction as it enables models to handle data from previously unseen domains. We introduce a novel DG method called Adversarial Intensity Attack (AdverIN), which leverages adversarial training to generate training data with an infinite number of styles.
  arXiv  Detail & Related papers  (2023-04-05T19:40:51Z)
• Improving Generalization with Domain Convex Game [32.07275105040802]
  Domain generalization tends to alleviate the poor generalization capability of deep neural networks by learning model with multiple source domains. A classical solution to DG is domain augmentation, the common belief of which is that diversifying source domains will be conducive to the out-of-distribution generalization. Our explorations reveal that the correlation between model generalization and the diversity of domains may be not strictly positive, which limits the effectiveness of domain augmentation.
  arXiv  Detail & Related papers  (2023-03-23T14:27:49Z)
• Normalization Perturbation: A Simple Domain Generalization Method for Real-World Domain Shifts [133.99270341855728]
  Real-world domain styles can vary substantially due to environment changes and sensor noises. Deep models only know the training domain style. We propose Normalization Perturbation to overcome this domain style overfitting problem.
  arXiv  Detail & Related papers  (2022-11-08T17:36:49Z)
• On-Device Domain Generalization [93.79736882489982]
  Domain generalization is critical to on-device machine learning applications. We find that knowledge distillation is a strong candidate for solving the problem. We propose a simple idea called out-of-distribution knowledge distillation (OKD), which aims to teach the student how the teacher handles (synthetic) out-of-distribution data.
  arXiv  Detail & Related papers  (2022-09-15T17:59:31Z)
• Federated and Generalized Person Re-identification through Domain and Feature Hallucinating [88.77196261300699]
  We study the problem of federated domain generalization (FedDG) for person re-identification (re-ID) We propose a novel method, called "Domain and Feature Hallucinating (DFH)", to produce diverse features for learning generalized local and global models. Our method achieves the state-of-the-art performance for FedDG on four large-scale re-ID benchmarks.
  arXiv  Detail & Related papers  (2022-03-05T09:15:13Z)
• Learning to Learn Single Domain Generalization [18.72451358284104]
  We propose a new method named adversarial domain augmentation to solve this Out-of-Distribution (OOD) generalization problem. The key idea is to leverage adversarial training to create "fictitious" yet "challenging" populations. To facilitate fast and desirable domain augmentation, we cast the model training in a meta-learning scheme and use a Wasserstein Auto-Encoder (WAE) to relax the widely used worst-case constraint.
  arXiv  Detail & Related papers  (2020-03-30T04:39:53Z)

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.