ترغب بنشر مسار تعليمي؟ اضغط هنا

Adversarial Feature Desensitization

76   0   0.0 ( 0 )
 نشر من قبل Pouya Bashivan
 تاريخ النشر 2020
والبحث باللغة English




اسأل ChatGPT حول البحث

Neural networks are known to be vulnerable to adversarial attacks -- slight but carefully constructed perturbations of the inputs which can drastically impair the networks performance. Many defense methods have been proposed for improving robustness of deep networks by training them on adversarially perturbed inputs. However, these models often remain vulnerable to new types of attacks not seen during training, and even to slightly strong



قيم البحث

اقرأ أيضاً

120 - Tao Bai , Jinnan Chen , Jun Zhao 2020
Deep learning models are shown to be vulnerable to adversarial examples. Though adversarial training can enhance model robustness, typical approaches are computationally expensive. Recent works proposed to transfer the robustness to adversarial attac ks across different tasks or models with soft labels.Compared to soft labels, feature contains rich semantic information and holds the potential to be applied to different downstream tasks. In this paper, we propose a novel approach called Guided Adversarial Contrastive Distillation (GACD), to effectively transfer adversarial robustness from teacher to student with features. We first formulate this objective as contrastive learning and connect it with mutual information. With a well-trained teacher model as an anchor, students are expected to extract features similar to the teacher. Then considering the potential errors made by teachers, we propose sample reweighted estimation to eliminate the negative effects from teachers. With GACD, the student not only learns to extract robust features, but also captures structural knowledge from the teacher. By extensive experiments evaluating over popular datasets such as CIFAR-10, CIFAR-100 and STL-10, we demonstrate that our approach can effectively transfer robustness across different models and even different tasks, and achieve comparable or better results than existing methods. Besides, we provide a detailed analysis of various methods, showing that students produced by our approach capture more structural knowledge from teachers and learn more robust features under adversarial attacks.
Learning domain-invariant representation is a dominant approach for domain generalization (DG), where we need to build a classifier that is robust toward domain shifts. However, previous domain-invariance-based methods overlooked the underlying depen dency of classes on domains, which is responsible for the trade-off between classification accuracy and domain invariance. Because the primary purpose of DG is to classify unseen domains rather than the invariance itself, the improvement of the invariance can negatively affect DG performance under this trade-off. To overcome the problem, this study first expands the analysis of the trade-off by Xie et. al., and provides the notion of accuracy-constrained domain invariance, which means the maximum domain invariance within a range that does not interfere with accuracy. We then propose a novel method adversarial feature learning with accuracy constraint (AFLAC), which explicitly leads to that invariance on adversarial training. Empirical validations show that the performance of AFLAC is superior to that of domain-invariance-based methods on both synthetic and three real-world datasets, supporting the importance of considering the dependency and the efficacy of the proposed method.
Missing value imputation is a challenging and well-researched topic in data mining. In this paper, we propose IFGAN, a missing value imputation algorithm based on Feature-specific Generative Adversarial Networks (GAN). Our idea is intuitive yet effec tive: a feature-specific generator is trained to impute missing values, while a discriminator is expected to distinguish the imputed values from observed ones. The proposed architecture is capable of handling different data types, data distributions, missing mechanisms, and missing rates. It also improves post-imputation analysis by preserving inter-feature correlations. We empirically show on several real-life datasets that IFGAN outperforms current state-of-the-art algorithm under various missing conditions.
We propose a new adversarial attack to Deep Neural Networks for image classification. Different from most existing attacks that directly perturb input pixels, our attack focuses on perturbing abstract features, more specifically, features that denote styles, including interpretable styles such as vivid colors and sharp outlines, and uninterpretable ones. It induces model misclassfication by injecting imperceptible style changes through an optimization procedure. We show that our attack can generate adversarial samples that are more natural-looking than the state-of-the-art unbounded attacks. The experiment also supports that existing pixel-space adversarial attack detection and defense techniques can hardly ensure robustness in the style related feature space.
We introduce supervised feature ranking and feature subset selection algorithms for multivariate time series (MTS) classification. Unlike most existing supervised/unsupervised feature selection algorithms for MTS our techniques do not require a featu re extraction step to generate a one-dimensional feature vector from the time series. Instead it is based on directly computing similarity between individual time series and assessing how well the resulting cluster structure matches the labels. The techniques are amenable to heterogeneous MTS data, where the time series measurements may have different sampling resolutions, and to multi-modal data.

الأسئلة المقترحة

التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا