Static prediction games for adversarial learning problems

By Michael Brückner, Christian Kanzow, Tobias Scheffer. 2012

216 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Safety, Robustness & Interpretability, Game & Decision Theory

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Safety, Robustness & Interpretability

In Game & Decision Theory

Citations in the corpus

Adversarial Example Games (2020, neutral): “Adversarial prediction games between a learner and a data generator have also been studied in the literature [11], and in certain situations correspond to a Stackelberg game Brückner and Scheffer [10].”
Wild Patterns: Ten Years After the Rise of Adversarial Machine Learning (2017, background): “Then, more rigorous approaches have introduced Nash and Stackelberg games for secure learning, deriving formal conditions for existence and uniqueness of the game equilibrium, under the assumption that each player knows everything about the opponents and the game [44, 85].”
Certified Defenses for Data Poisoning Attacks (2017, methodology): “Bilevel optimization has been considered in the related setting of Stackelberg games (Brückner and Scheffer, 2011; Brückner et al., 2012; Zhou and Kantarcioglu, 2016), and is natural to apply here as well.”
Evasion Attacks against Machine Learning at Test Time (2013, methodology): “Under certain conditions, such problems can be solved using a Nash equilibrium approach [7, 8].”
Is Feature Selection Secure against Training Data Poisoning? (2015, methodology): “The worst-case scenario in which the attacker has full knowledge of the attacked system, is usually referred to as perfect knowledge case (Biggio et al., 2014; 2013b; 2012; Kloft & Laskov, 2010; Brückner et al., 2012; Huang et al., 2011; Barreno et al., 2006), and it allows one to empirically evalua”
Practical Evasion of a Learning-Based Classifier: A Case Study (2014, methodology): “In this paper, we experimentally investigate the effectiveness of classifier evasion using a real, deployed system, PDFRATE, as a test case.”
Evasion and Hardening of Tree Ensemble Classifiers (2015, background): “…considering the evasion of linear or otherwise differentiable models (Dalvi et al., 2004; L., 2005; Lowd & Meek, 2005; Nelson et al., 2012; Br¨uckner et al., 2012; Fawzi et al., 2014; Biggio et al., 2013; Szegedy et al., 2013; Srndic & Laskov, 2014), we are only aware of a single paper…”
Poisoning behavioral malware clustering (2014, background): “The worst-case scenario in which the attacker has full knowledge of the targeted system is usually referred to as perfect knowledge [2, 5–8, 10, 17, 22].”
Adversarial Detection of Flash Malware: Limitations and Open Issues (2017, background): “], [13].”
Evading Classifiers by Morphing in the Dark (2017, background): “Various studies [7, 11, 12] suggest that they are likely susceptible to evasion”
Yes, Machine Learning Can Be More Secure! A Case Study on Android Malware Detection (2017, methodology): “puted between the source malware data and its manipulated versions [7], [15], [19], [21], [29], [40].”
Randomized Prediction Games for Adversarial Machine Learning (2016, methodology): “Any solution θ to this problem can be shown to correspond bijectively to a Nash equilibrium of a game having gr as pseudogradient [11], [32].”
Adversarial Feature Selection Against Evasion Attacks (2016, background): “assumption is, however, likely to be violated in adversarial settings, since attackers may carefully manipulate the input data to downgrade the system’s performance [8], [10]–[12], [39].”
Security Evaluation of Support Vector Machines in Adversarial Environments (2014, background): “Apart from application-specific work, several theoretical models of adversarial learning have been proposed [4, 17, 26, 36, 40, 42, 46, 53].”
Is data clustering in adversarial settings secure? (2013, methodology): “This choice allows us to bound the divergence between the original target samples in A and the manipulated ones, as typically done in adversarial learning [20, 17, 8, 6].”
Support vector machines under adversarial label contamination (2015, background): “Different game formulations can also be exploited if the players use non-antagonistic objective functions, as in [22].”
Data Poisoning Attacks against Autoregressive Models (2016, background): “A somewhat orthogonal line of research has posed the problem of learning in the presence of adversaries in a game theoretic context (Liu and Chawla 2009; Br¨uckner, Kanzow, and Scheffer 2012; Dalvi et al. 2004; Br¨uckner and Scheffer 2009; 2011).”
Towards Making Systems Forget with Machine Unlearning (2015, neutral): “[28] model the learner and the data-pollution attacker as a game and prove that the game has a unique Nash equilibrium.”
Feature Conservation in Adversarial Classifier Evasion: A Case Study (2017, background): “…leveraging traditional machine learning techniques, such as regularization [35], [36], [23], [28], to game theoretic approaches in which the defender chooses a classiﬁer that the attacker attempts to evade [7], [4], [18], [39], [5], to retraining with adversarial evasion examples [19], [15], [16].”