Consequential decision-making typically incentivizes individuals to behave strategically, tailoring their behavior to the specifics of the decision rule. A long line of work has therefore sought to counteract strategic behavior by designing more conservative decision boundaries in an effort to increase robustness to the effects of strategic covariate shift. We show that these efforts benefit the institutional decision maker at the expense of the individuals being classified. Introducing a notion of social burden, we prove that any increase in institutional utility necessarily leads to a corresponding increase in social burden. Moreover, we show that the negative externalities of strategic classification can disproportionately harm disadvantaged groups in the population. Our results highlight that strategy-robustness must be weighed against considerations of social welfare and fairness.
{"title":"The Social Cost of Strategic Classification","authors":"S. Milli, John Miller, A. Dragan, Moritz Hardt","doi":"10.1145/3287560.3287576","DOIUrl":"https://doi.org/10.1145/3287560.3287576","url":null,"abstract":"Consequential decision-making typically incentivizes individuals to behave strategically, tailoring their behavior to the specifics of the decision rule. A long line of work has therefore sought to counteract strategic behavior by designing more conservative decision boundaries in an effort to increase robustness to the effects of strategic covariate shift. We show that these efforts benefit the institutional decision maker at the expense of the individuals being classified. Introducing a notion of social burden, we prove that any increase in institutional utility necessarily leads to a corresponding increase in social burden. Moreover, we show that the negative externalities of strategic classification can disproportionately harm disadvantaged groups in the population. Our results highlight that strategy-robustness must be weighed against considerations of social welfare and fairness.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":"45 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2018-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90979188","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Michael Kearns, S. Neel, Aaron Roth, Zhiwei Steven Wu
Kearns, Neel, Roth, and Wu [ICML 2018] recently proposed a notion of rich subgroup fairness intended to bridge the gap between statistical and individual notions of fairness. Rich subgroup fairness picks a statistical fairness constraint (say, equalizing false positive rates across protected groups), but then asks that this constraint hold over an exponentially or infinitely large collection of subgroups defined by a class of functions with bounded VC dimension. They give an algorithm guaranteed to learn subject to this constraint, under the condition that it has access to oracles for perfectly learning absent a fairness constraint. In this paper, we undertake an extensive empirical evaluation of the algorithm of Kearns et al. On four real datasets for which fairness is a concern, we investigate the basic convergence of the algorithm when instantiated with fast heuristics in place of learning oracles, measure the tradeoffs between fairness and accuracy, and compare this approach with the recent algorithm of Agarwal, Beygelzeimer, Dudik, Langford, and Wallach [ICML 2018], which implements weaker and more traditional marginal fairness constraints defined by individual protected attributes. We find that in general, the Kearns et al. algorithm converges quickly, large gains in fairness can be obtained with mild costs to accuracy, and that optimizing accuracy subject only to marginal fairness leads to classifiers with substantial subgroup unfairness. We also provide a number of analyses and visualizations of the dynamics and behavior of the Kearns et al. algorithm. Overall we find this algorithm to be effective on real data, and rich subgroup fairness to be a viable notion in practice.
{"title":"An Empirical Study of Rich Subgroup Fairness for Machine Learning","authors":"Michael Kearns, S. Neel, Aaron Roth, Zhiwei Steven Wu","doi":"10.1145/3287560.3287592","DOIUrl":"https://doi.org/10.1145/3287560.3287592","url":null,"abstract":"Kearns, Neel, Roth, and Wu [ICML 2018] recently proposed a notion of rich subgroup fairness intended to bridge the gap between statistical and individual notions of fairness. Rich subgroup fairness picks a statistical fairness constraint (say, equalizing false positive rates across protected groups), but then asks that this constraint hold over an exponentially or infinitely large collection of subgroups defined by a class of functions with bounded VC dimension. They give an algorithm guaranteed to learn subject to this constraint, under the condition that it has access to oracles for perfectly learning absent a fairness constraint. In this paper, we undertake an extensive empirical evaluation of the algorithm of Kearns et al. On four real datasets for which fairness is a concern, we investigate the basic convergence of the algorithm when instantiated with fast heuristics in place of learning oracles, measure the tradeoffs between fairness and accuracy, and compare this approach with the recent algorithm of Agarwal, Beygelzeimer, Dudik, Langford, and Wallach [ICML 2018], which implements weaker and more traditional marginal fairness constraints defined by individual protected attributes. We find that in general, the Kearns et al. algorithm converges quickly, large gains in fairness can be obtained with mild costs to accuracy, and that optimizing accuracy subject only to marginal fairness leads to classifiers with substantial subgroup unfairness. We also provide a number of analyses and visualizations of the dynamics and behavior of the Kearns et al. algorithm. Overall we find this algorithm to be effective on real data, and rich subgroup fairness to be a viable notion in practice.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":"6 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2018-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82275660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We show through theory and experiment that gradient-based explanations of a model quickly reveal the model itself. Our results speak to a tension between the desire to keep a proprietary model secret and the ability to offer model explanations. On the theoretical side, we give an algorithm that provably learns a two-layer ReLU network in a setting where the algorithm may query the gradient of the model with respect to chosen inputs. The number of queries is independent of the dimension and nearly optimal in its dependence on the model size. Of interest not only from a learning-theoretic perspective, this result highlights the power of gradients rather than labels as a learning primitive. Complementing our theory, we give effective heuristics for reconstructing models from gradient explanations that are orders of magnitude more query-efficient than reconstruction attacks relying on prediction interfaces.
{"title":"Model Reconstruction from Model Explanations","authors":"S. Milli, Ludwig Schmidt, A. Dragan, Moritz Hardt","doi":"10.1145/3287560.3287562","DOIUrl":"https://doi.org/10.1145/3287560.3287562","url":null,"abstract":"We show through theory and experiment that gradient-based explanations of a model quickly reveal the model itself. Our results speak to a tension between the desire to keep a proprietary model secret and the ability to offer model explanations. On the theoretical side, we give an algorithm that provably learns a two-layer ReLU network in a setting where the algorithm may query the gradient of the model with respect to chosen inputs. The number of queries is independent of the dimension and nearly optimal in its dependence on the model size. Of interest not only from a learning-theoretic perspective, this result highlights the power of gradients rather than labels as a learning primitive. Complementing our theory, we give effective heuristics for reconstructing models from gradient explanations that are orders of magnitude more query-efficient than reconstruction attacks relying on prediction interfaces.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":"25 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2018-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84483553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
L. E. Celis, Lingxiao Huang, Vijay Keswani, Nisheeth K. Vishnoi
Developing classification algorithms that are fair with respect to sensitive attributes of the data is an important problem due to the increased deployment of classification algorithms in societal contexts. Several recent works have focused on studying classification with respect to specific fairness metrics, modeled the corresponding fair classification problem as constrained optimization problems, and developed tailored algorithms to solve them. Despite this, there still remain important metrics for which there are no fair classifiers with theoretical guarantees; primarily because the resulting optimization problem is non-convex. The main contribution of this paper is a meta-algorithm for classification that can take as input a general class of fairness constraints with respect to multiple non-disjoint and multi-valued sensitive attributes, and which comes with provable guarantees. In particular, our algorithm can handle non-convex "linear fractional" constraints (which includes fairness constraints such as predictive parity) for which no prior algorithm was known. Key to our results is an algorithm for a family of classification problems with convex constraints along with a reduction from classification problems with linear fractional constraints to this family. Empirically, we observe that our algorithm is fast, can achieve near-perfect fairness with respect to various fairness metrics, and the loss in accuracy due to the imposed fairness constraints is often small.
{"title":"Classification with Fairness Constraints: A Meta-Algorithm with Provable Guarantees","authors":"L. E. Celis, Lingxiao Huang, Vijay Keswani, Nisheeth K. Vishnoi","doi":"10.1145/3287560.3287586","DOIUrl":"https://doi.org/10.1145/3287560.3287586","url":null,"abstract":"Developing classification algorithms that are fair with respect to sensitive attributes of the data is an important problem due to the increased deployment of classification algorithms in societal contexts. Several recent works have focused on studying classification with respect to specific fairness metrics, modeled the corresponding fair classification problem as constrained optimization problems, and developed tailored algorithms to solve them. Despite this, there still remain important metrics for which there are no fair classifiers with theoretical guarantees; primarily because the resulting optimization problem is non-convex. The main contribution of this paper is a meta-algorithm for classification that can take as input a general class of fairness constraints with respect to multiple non-disjoint and multi-valued sensitive attributes, and which comes with provable guarantees. In particular, our algorithm can handle non-convex \"linear fractional\" constraints (which includes fairness constraints such as predictive parity) for which no prior algorithm was known. Key to our results is an algorithm for a family of classification problems with convex constraints along with a reduction from classification problems with linear fractional constraints to this family. Empirically, we observe that our algorithm is fast, can achieve near-perfect fairness with respect to various fairness metrics, and the loss in accuracy due to the imposed fairness constraints is often small.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":"50 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2018-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80526416","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sorelle A. Friedler, C. Scheidegger, Suresh Venkatasubramanian, Sonam Choudhary, Evan P. Hamilton, Derek Roth
Computers are increasingly used to make decisions that have significant impact on people's lives. Often, these predictions can affect different population subgroups disproportionately. As a result, the issue of fairness has received much recent interest, and a number of fairness-enhanced classifiers have appeared in the literature. This paper seeks to study the following questions: how do these different techniques fundamentally compare to one another, and what accounts for the differences? Specifically, we seek to bring attention to many under-appreciated aspects of such fairness-enhancing interventions that require investigation for these algorithms to receive broad adoption. We present the results of an open benchmark we have developed that lets us compare a number of different algorithms under a variety of fairness measures and existing datasets. We find that although different algorithms tend to prefer specific formulations of fairness preservations, many of these measures strongly correlate with one another. In addition, we find that fairness-preserving algorithms tend to be sensitive to fluctuations in dataset composition (simulated in our benchmark by varying training-test splits) and to different forms of preprocessing, indicating that fairness interventions might be more brittle than previously thought.
{"title":"A comparative study of fairness-enhancing interventions in machine learning","authors":"Sorelle A. Friedler, C. Scheidegger, Suresh Venkatasubramanian, Sonam Choudhary, Evan P. Hamilton, Derek Roth","doi":"10.1145/3287560.3287589","DOIUrl":"https://doi.org/10.1145/3287560.3287589","url":null,"abstract":"Computers are increasingly used to make decisions that have significant impact on people's lives. Often, these predictions can affect different population subgroups disproportionately. As a result, the issue of fairness has received much recent interest, and a number of fairness-enhanced classifiers have appeared in the literature. This paper seeks to study the following questions: how do these different techniques fundamentally compare to one another, and what accounts for the differences? Specifically, we seek to bring attention to many under-appreciated aspects of such fairness-enhancing interventions that require investigation for these algorithms to receive broad adoption. We present the results of an open benchmark we have developed that lets us compare a number of different algorithms under a variety of fairness measures and existing datasets. We find that although different algorithms tend to prefer specific formulations of fairness preservations, many of these measures strongly correlate with one another. In addition, we find that fairness-preserving algorithms tend to be sensitive to fluctuations in dataset composition (simulated in our benchmark by varying training-test splits) and to different forms of preprocessing, indicating that fairness interventions might be more brittle than previously thought.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2018-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81073706","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：