Explaining and Harnessing Adversarial Examples
时间: 2024-04-12 17:32:51 浏览: 16
对抗性样本是指对于机器学习模型来说,经过有意设计的、微小的扰动,能够使得该模型的输出结果产生错误。解释和利用对抗性样本是指尝试理解对抗性样本对机器学习模型产生影响的原因,并利用这些知识来提高模型的鲁棒性,使得模型在面对对抗性样本时表现更加可靠。具体而言,可以通过研究对抗性样本的生成方法和性质,开发更加健壮的机器学习算法或者防御机制,提高模型的鲁棒性。
相关问题
explaining and harnessing adversarial examples
对抗性样本是指对于机器学习模型来说,经过有意设计的、微小的扰动,能够使得该模型的输出结果产生错误。解释和利用对抗性样本是指尝试理解对抗性样本对机器学习模型产生影响的原因,并利用这些知识来提高模型的鲁棒性,使得模型在面对对抗性样本时表现更加可靠。具体而言,可以通过研究对抗性样本的生成方法和性质,开发更加健壮的机器学习算法或者防御机制,提高模型的鲁棒性。
对抗神经网络的文献有哪些
以下是对抗神经网络的文献:
1. Goodfellow, I., Shlens, J., & Szegedy, C. (2014). Explaining and Harnessing Adversarial Examples.
2. Papernot, N., McDaniel, P., Jha, S., Fredrikson, M., Celik, Z. B., & Swami, A. (2016). The limitations of deep learning in adversarial settings.
3. Kurakin, A., Goodfellow, I., & Bengio, S. (2016). Adversarial examples in the physical world.
4. Szegedy, C., Zaremba, W., Sutskever, I., Bruna, J., Erhan, D., Goodfellow, I., & Fergus, R. (2013). Intriguing properties of neural networks.
5. Carlini, N., & Wagner, D. (2017). Towards evaluating the robustness of neural networks.
6. Nguyen, A., Yosinski, J., & Clune, J. (2015). Deep neural networks are easily fooled: High confidence predictions for unrecognizable images.
7. Madry, A., Makelov, A., Schmidt, L., Tsipras, D., & Vladu, A. (2018). Towards deep learning models resistant to adversarial attacks.
8. Brown, T. B., Mané, D., Roy, A., Abadi, M., & Gilmer, J. (2017). Adversarial patch.
9. Moosavi-Dezfooli, S. M., Fawzi, A., & Frossard, P. (2016). Deepfool: a simple and accurate method to fool deep neural networks.
10. Huang, X., Li, Y., Poursaeed, O., Hopcroft, J., & Belongie, S. (2017). Adversarial attacks on neural network policies.