Strategic Classification

Hardt, Moritz; Megiddo, Nimrod; Papadimitriou, Christos H.; Wootters, Mary

doi:10.1145/2840728.2840730

Cited by 209 publications

(228 citation statements)

References 10 publications

Supporting

Mentioning

222

Contrasting

Order By: Relevance

“…This is a crucial distinction in highstakes decision making as different error types present asymmetric incentives for individuals, as explained in Section 2; for example, a high false positive rate in hiring would encourage underqualified job applicants. Hu et al [2019] and Milli et al [2019] study the disparate impact of being robust towards strategic manipulation [see e.g., Hardt et al, 2016a], where individuals respond to machine learning systems by manipulating their features to get a better classification. In contrast to our model ( Figure 5), their setting models the individual as intervening directly on their features, X, and this is assumed to have no effect on their qualification Y .…”

Section: Related Workmentioning

confidence: 99%

The disparate equilibria of algorithmic decision making when individuals invest rationally

Liu

Wilson

Haghtalab

et al. 2020

Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency

View full text Add to dashboard Cite

The long-term impact of algorithmic decision making is shaped by the dynamics between the deployed decision rule and individuals' response. Focusing on settings where each individual desires a positive classification-including many important applications such as hiring and school admissions, we study a dynamic learning setting where individuals invest in a positive outcome based on their group's expected gain and the decision rule is updated to maximize institutional benefit. By characterizing the equilibria of these dynamics, we show that natural challenges to desirable long-term outcomes arise due to heterogeneity across groups and the lack of realizability. We consider two interventions, decoupling the decision rule by group and subsidizing the cost of investment. We show that decoupling achieves optimal outcomes in the realizable case but has discrepant effects that may depend on the initial conditions otherwise. In contrast, subsidizing the cost of investment is shown to create better equilibria for the disadvantaged group even in the absence of realizability.

show abstract

Section: Related Workmentioning

confidence: 99%

The disparate equilibria of algorithmic decision making when individuals invest rationally

Liu

Wilson

Haghtalab

et al. 2020

Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency

View full text Add to dashboard Cite

show abstract

“…Adversary Attack Model Validation domain with adversary Zero Sum Games [11] Full information about learner's Causative attacks Spam filtering utility, cost and classifier parameters [35,39,31,57] No information about learner's Exploratory attacks by changing Spam filtering utility, costs and classifier parameters values of future input [17,12,52] No information about learner's Exploratory attacks by removing Spam filtering utility, costs and classifier parameters features from future input [23] With and without info. about prob.…”

Section: Initial Information About Learnermentioning

confidence: 99%

A Survey of Game Theoretic Approaches for Adversarial Machine Learning in Cybersecurity Tasks

Dasgupta¹,

Collins²

2019

AI Magazine

View full text Add to dashboard Cite

Machine learning techniques are currently used extensively for automating various cybersecurity tasks. Most of these techniques utilize supervised learning algorithms that rely on training the algorithm to classify incoming data into different categories, using data encountered in the relevant domain. A critical vulnerability of these algorithms is that they are susceptible to adversarial attacks where a malicious entity called an adversary deliberately alters the training data to misguide the learning algorithm into making classification errors. Adversarial attacks could render the learning algorithm unsuitable to use and leave critical systems vulnerable to cybersecurity attacks. Our paper provides a detailed survey of the state-of-the-art techniques that are used to make a machine learning algorithm robust against adversarial attacks using the computational framework of game theory. We also discuss open problems and challenges and possible directions for further research that would make deep machine learning-based systems more robust and reliable for cybersecurity tasks.

show abstract

“…In these works, there often is no cost for reporting one's data and the data analyst doesn't use monetary payments. These works attempt to design or identify mechanisms (inference or learning processes) that are robust to potential data manipulations [Dekel et al, 2010, Meir et al, 2011, Perote and Perote-Peña, 2003, Hardt et al, 2016, Dong et al, 2017, Chen et al, 2018b].…”

Section: Other Related Workmentioning

confidence: 99%

Prior-free Data Acquisition for Accurate Statistical Estimation

Zheng

2019

Proceedings of the 2019 ACM Conference on Economics and Computation

View full text Add to dashboard Cite

We study a data analyst's problem of acquiring data from self-interested individuals to obtain an accurate estimation of some statistic of a population, subject to an expected budget constraint. Each data holder incurs a cost, which is unknown to the data analyst, to acquire and report his data. The cost can be arbitrarily correlated with the data. The data analyst has an expected budget that she can use to incentivize individuals to provide their data. The goal is to design a joint acquisition-estimation mechanism to optimize the performance of the produced estimator, without any prior information on the underlying distribution of cost and data. We investigate two types of estimations: unbiased point estimation and confidence interval estimation.Unbiased estimators: We design a truthful, individually rational, online mechanism to acquire data from individuals and output an unbiased estimator of the population mean when the data analyst has no prior information on the cost-data distribution and individuals arrive in a random order. The performance of this mechanism matches that of the optimal mechanism, which knows the true cost distribution, within a constant factor. The performance of an estimator is evaluated by its variance under the worst-case cost-data correlation. Confidence intervals:We characterize an approximately optimal (within a factor 2) mechanism for obtaining a confidence interval of the population mean when the data analyst knows the true cost distribution at the beginning. This mechanism is efficiently computable. We then design a truthful, individually rational, online algorithm that is only worse than the approximately optimal mechanism by a constant factor. The performance of an estimator is evaluated by its expected length under the worst-case cost-data correlation.

show abstract

Strategic Classification

Cited by 209 publications

References 10 publications

The disparate equilibria of algorithmic decision making when individuals invest rationally

The disparate equilibria of algorithmic decision making when individuals invest rationally

A Survey of Game Theoretic Approaches for Adversarial Machine Learning in Cybersecurity Tasks

Prior-free Data Acquisition for Accurate Statistical Estimation

Contact Info

Product

Resources

About