[paper]ADVERSARIAL REPROGRAMMING OF NEURAL NETWORKS
傳統對抗樣本目的是使模型分類錯誤，本文通過則是使模型執行特定任務（攻擊者設定），且該任務可以未被訓練過。 We introduce attacks that instead reprogram the target model to perform a task chosen by the attacker—without the attacker needing to specify or co
相關文章
- 1. Neural Bag-of-Ngrams
- 2. ICLR2014_Intriguing properties of neural networks
- 3. Applications of Graph Neural Networks
- 4. Convolutional Neural Networks---Foundations of Convolutional Neural Networks
- 5. On the Number of Linear Regions of Deep Neural Networks
- 6. Deeplearning - Overview of Convolution Neural Network
- 7. DANN：Domain-Adversarial Training of Neural Networks
- 8. 018 Interpretability of Neural Networks (Lecture 7)
- 9. CS224W 18.1-Limitations of Graph Neural Network
- 10. A Neural Algorithm of Artistic Style
- 更多相關文章...
- • XSLT
- • XSLT
- • JDK13 GA發佈：5大特性解讀
- • 使用Rxjava計算圓周率
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. Neural Bag-of-Ngrams
- 2. ICLR2014_Intriguing properties of neural networks
- 3. Applications of Graph Neural Networks
- 4. Convolutional Neural Networks---Foundations of Convolutional Neural Networks
- 5. On the Number of Linear Regions of Deep Neural Networks
- 6. Deeplearning - Overview of Convolution Neural Network
- 7. DANN：Domain-Adversarial Training of Neural Networks
- 8. 018 Interpretability of Neural Networks (Lecture 7)
- 9. CS224W 18.1-Limitations of Graph Neural Network
- 10. A Neural Algorithm of Artistic Style