Learning with a Strong Adversary
نویسندگان
چکیده
In this paper, we propose a method, learning with adversary, to learn a robust network. Our method takes finding adversarial examples as its mediate step. A new and simple way of finding adversarial examples are presented and experimentally shown to be more ‘efficient’. Lastly, experimental results shows our learning method greatly improves the robustness of the learned network.
منابع مشابه
Three Strategies to Success: Learning Adversary Models in Security Games
State-of-the-art applications of Stackelberg security games — including wildlife protection — offer a wealth of data, which can be used to learn the behavior of the adversary. But existing approaches either make strong assumptions about the structure of the data, or gather new data through online algorithms that are likely to play severely suboptimal strategies. We develop a new approach to lea...
متن کاملLearning Adversary Behavior in Security Games: A PAC Model Perspective
Recent applications of Stackelberg Security Games (SSG), from wildlife crime to urban crime, have employed machine learning tools to learn and predict adversary behavior using available data about defender-adversary interactions. Given these recent developments, this paper commits to an approach of directly learning the response function of the adversary. Using the PAC model, this paper lays a ...
متن کاملPrivacy-Preserving Abuse Detection in Future Decentralised Online Social Networks
Future online social networks need to not only protect sensitive data of their users, but also protect them from abusive behavior coming from malicious participants in the network. We investigate the use of supervised learning techniques to detect abusive behavior and describe privacy-preserving protocols to compute the feature set required by abuse classification algorithms in a secure and pri...
متن کاملAdversarial Neural Machine Translation
In this paper, we study a new learning paradigm for Neural Machine Translation (NMT). Instead of maximizing the likelihood of the human translation as in previous works, we minimize the distinction between human translation and the translation given by a NMT model. To achieve this goal, inspired by the recent success of generative adversarial networks (GANs), we employ an adversarial training a...
متن کاملAdaptive Regret Minimization in Bounded-Memory Games
Online learning algorithms that minimize regret provide strong guarantees in situations that involve repeatedly making decisions in an uncertain environment, e.g. a driver deciding what route to drive to work every day. While regret minimization has been extensively studied in repeated games, we study regret minimization for a richer class of games called bounded memory games. In each round of ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- CoRR
دوره abs/1511.03034 شماره
صفحات -
تاریخ انتشار 2015