Recent work on interpretability in machine learning and AI has focused on the building of simplified models that approximate the true criteria used to make decisions. These models are a useful pedagogical device for teaching trained professionals how to predict what decisions will be made by the complex system, and most importantly how the system might break. However, when considering any such model it's important to remember Box's maxim that "All models are wrong but some are useful." We focus on the distinction between these models and explanations in philosophy and sociology. These models can be understood as a "do it yourself kit" for explanations, allowing a practitioner to directly answer "what if questions" or generate contrastive explanations without external assistance. Although a valuable ability, giving these models as explanations appears more difficult than necessary, and other forms of explanation may not have the same trade-offs. We contrast the different schools of thought on what makes an explanation, and suggest that machine learning might benefit from viewing the problem more broadly.
{"title":"Explaining Explanations in AI","authors":"B. Mittelstadt, Chris Russell, Sandra Wachter","doi":"10.1145/3287560.3287574","DOIUrl":"https://doi.org/10.1145/3287560.3287574","url":null,"abstract":"Recent work on interpretability in machine learning and AI has focused on the building of simplified models that approximate the true criteria used to make decisions. These models are a useful pedagogical device for teaching trained professionals how to predict what decisions will be made by the complex system, and most importantly how the system might break. However, when considering any such model it's important to remember Box's maxim that \"All models are wrong but some are useful.\" We focus on the distinction between these models and explanations in philosophy and sociology. These models can be understood as a \"do it yourself kit\" for explanations, allowing a practitioner to directly answer \"what if questions\" or generate contrastive explanations without external assistance. Although a valuable ability, giving these models as explanations appears more difficult than necessary, and other forms of explanation may not have the same trade-offs. We contrast the different schools of thought on what makes an explanation, and suggest that machine learning might benefit from viewing the problem more broadly.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77630728","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}
Margaret Mitchell, Simone Wu, Andrew Zaldivar, Parker Barnes, Lucy Vasserman, B. Hutchinson, Elena Spitzer, Inioluwa Deborah Raji, Timnit Gebru
Trained machine learning models are increasingly used to perform high-impact tasks in areas such as law enforcement, medicine, education, and employment. In order to clarify the intended use cases of machine learning models and minimize their usage in contexts for which they are not well suited, we recommend that released models be accompanied by documentation detailing their performance characteristics. In this paper, we propose a framework that we call model cards, to encourage such transparent model reporting. Model cards are short documents accompanying trained machine learning models that provide benchmarked evaluation in a variety of conditions, such as across different cultural, demographic, or phenotypic groups (e.g., race, geographic location, sex, Fitzpatrick skin type [15]) and intersectional groups (e.g., age and race, or sex and Fitzpatrick skin type) that are relevant to the intended application domains. Model cards also disclose the context in which models are intended to be used, details of the performance evaluation procedures, and other relevant information. While we focus primarily on human-centered machine learning models in the application fields of computer vision and natural language processing, this framework can be used to document any trained machine learning model. To solidify the concept, we provide cards for two supervised models: One trained to detect smiling faces in images, and one trained to detect toxic comments in text. We propose model cards as a step towards the responsible democratization of machine learning and related artificial intelligence technology, increasing transparency into how well artificial intelligence technology works. We hope this work encourages those releasing trained machine learning models to accompany model releases with similar detailed evaluation numbers and other relevant documentation.
{"title":"Model Cards for Model Reporting","authors":"Margaret Mitchell, Simone Wu, Andrew Zaldivar, Parker Barnes, Lucy Vasserman, B. Hutchinson, Elena Spitzer, Inioluwa Deborah Raji, Timnit Gebru","doi":"10.1145/3287560.3287596","DOIUrl":"https://doi.org/10.1145/3287560.3287596","url":null,"abstract":"Trained machine learning models are increasingly used to perform high-impact tasks in areas such as law enforcement, medicine, education, and employment. In order to clarify the intended use cases of machine learning models and minimize their usage in contexts for which they are not well suited, we recommend that released models be accompanied by documentation detailing their performance characteristics. In this paper, we propose a framework that we call model cards, to encourage such transparent model reporting. Model cards are short documents accompanying trained machine learning models that provide benchmarked evaluation in a variety of conditions, such as across different cultural, demographic, or phenotypic groups (e.g., race, geographic location, sex, Fitzpatrick skin type [15]) and intersectional groups (e.g., age and race, or sex and Fitzpatrick skin type) that are relevant to the intended application domains. Model cards also disclose the context in which models are intended to be used, details of the performance evaluation procedures, and other relevant information. While we focus primarily on human-centered machine learning models in the application fields of computer vision and natural language processing, this framework can be used to document any trained machine learning model. To solidify the concept, we provide cards for two supervised models: One trained to detect smiling faces in images, and one trained to detect toxic comments in text. We propose model cards as a step towards the responsible democratization of machine learning and related artificial intelligence technology, increasing transparency into how well artificial intelligence technology works. We hope this work encourages those releasing trained machine learning models to accompany model releases with similar detailed evaluation numbers and other relevant documentation.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90690620","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}
R. Canetti, A. Cohen, Nishanth Dikkala, Govind Ramnarayan, Sarah Scheffler, Adam D. Smith
A popular methodology for building binary decision-making classifiers in the presence of imperfect information is to first construct a calibrated non-binary "scoring" classifier, and then to post-process this score to obtain a binary decision. We study various fairness (or, error-balance) properties of this methodology, when the non-binary scores are calibrated over all protected groups, and with a variety of post-processing algorithms. Specifically, we show: First, there does not exist a general way to post-process a calibrated classifier to equalize protected groups' positive or negative predictive value (PPV or NPV). For certain "nice" calibrated classifiers, either PPV or NPV can be equalized when the post-processor uses different thresholds across protected groups. Still, when the post-processing consists of a single global threshold across all groups, natural fairness properties, such as equalizing PPV in a nontrivial way, do not hold even for "nice" classifiers. Second, when the post-processing stage is allowed to defer on some decisions (that is, to avoid making a decision by handing off some examples to a separate process), then for the non-deferred decisions, the resulting classifier can be made to equalize PPV, NPV, false positive rate (FPR) and false negative rate (FNR) across the protected groups. This suggests a way to partially evade the impossibility results of Chouldechova and Kleinberg et al., which preclude equalizing all of these measures simultaneously. We also present different deferring strategies and show how they affect the fairness properties of the overall system. We evaluate our post-processing techniques using the COMPAS data set from 2016.
{"title":"From Soft Classifiers to Hard Decisions: How fair can we be?","authors":"R. Canetti, A. Cohen, Nishanth Dikkala, Govind Ramnarayan, Sarah Scheffler, Adam D. Smith","doi":"10.1145/3287560.3287561","DOIUrl":"https://doi.org/10.1145/3287560.3287561","url":null,"abstract":"A popular methodology for building binary decision-making classifiers in the presence of imperfect information is to first construct a calibrated non-binary \"scoring\" classifier, and then to post-process this score to obtain a binary decision. We study various fairness (or, error-balance) properties of this methodology, when the non-binary scores are calibrated over all protected groups, and with a variety of post-processing algorithms. Specifically, we show: First, there does not exist a general way to post-process a calibrated classifier to equalize protected groups' positive or negative predictive value (PPV or NPV). For certain \"nice\" calibrated classifiers, either PPV or NPV can be equalized when the post-processor uses different thresholds across protected groups. Still, when the post-processing consists of a single global threshold across all groups, natural fairness properties, such as equalizing PPV in a nontrivial way, do not hold even for \"nice\" classifiers. Second, when the post-processing stage is allowed to defer on some decisions (that is, to avoid making a decision by handing off some examples to a separate process), then for the non-deferred decisions, the resulting classifier can be made to equalize PPV, NPV, false positive rate (FPR) and false negative rate (FNR) across the protected groups. This suggests a way to partially evade the impossibility results of Chouldechova and Kleinberg et al., which preclude equalizing all of these measures simultaneously. We also present different deferring strategies and show how they affect the fairness properties of the overall system. We evaluate our post-processing techniques using the COMPAS data set from 2016.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82005712","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}
Classification models are often used to make decisions that affect humans: whether to approve a loan application, extend a job offer, or provide insurance. In such applications, individuals should have the ability to change the decision of the model. When a person is denied a loan by a credit scoring model, for example, they should be able to change the input variables of the model in a way that will guarantee approval. Otherwise, this person will be denied the loan so long as the model is deployed, and -- more importantly --will lack agency over a decision that affects their livelihood. In this paper, we propose to evaluate a linear classification model in terms of recourse, which we define as the ability of a person to change the decision of the model through actionable input variables (e.g., income vs. age or marital status). We present an integer programming toolkit to: (i) measure the feasibility and difficulty of recourse in a target population; and (ii) generate a list of actionable changes for a person to obtain a desired outcome. We discuss how our tools can inform different stakeholders by using them to audit recourse for credit scoring models built with real-world datasets. Our results illustrate how recourse can be significantly affected by common modeling practices, and motivate the need to evaluate recourse in algorithmic decision-making.
{"title":"Actionable Recourse in Linear Classification","authors":"Berk Ustun, Alexander Spangher, Yang Liu","doi":"10.1145/3287560.3287566","DOIUrl":"https://doi.org/10.1145/3287560.3287566","url":null,"abstract":"Classification models are often used to make decisions that affect humans: whether to approve a loan application, extend a job offer, or provide insurance. In such applications, individuals should have the ability to change the decision of the model. When a person is denied a loan by a credit scoring model, for example, they should be able to change the input variables of the model in a way that will guarantee approval. Otherwise, this person will be denied the loan so long as the model is deployed, and -- more importantly --will lack agency over a decision that affects their livelihood. In this paper, we propose to evaluate a linear classification model in terms of recourse, which we define as the ability of a person to change the decision of the model through actionable input variables (e.g., income vs. age or marital status). We present an integer programming toolkit to: (i) measure the feasibility and difficulty of recourse in a target population; and (ii) generate a list of actionable changes for a person to obtain a desired outcome. We discuss how our tools can inform different stakeholders by using them to audit recourse for credit scoring models built with real-world datasets. Our results illustrate how recourse can be significantly affected by common modeling practices, and motivate the need to evaluate recourse in algorithmic decision-making.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83467159","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 identify and explore differential access to population-level signaling (also known as information design) as a source of unequal access to opportunity. A population-level signaler has potentially noisy observations of a binary type for each member of a population and, based on this, produces a signal about each member. A decision-maker infers types from signals and accepts those individuals whose type is high in expectation. We assume the signaler of the disadvantaged population reveals her observations to the decision-maker, whereas the signaler of the advantaged population forms signals strategically. We study the expected utility of the populations as measured by the fraction of accepted members, as well as the false positive rates (FPR) and false negative rates (FNR). We first show the intuitive results that for a fixed environment, the advantaged population has higher expected utility, higher FPR, and lower FNR, than the disadvantaged one (despite having identical population quality), and that more accurate observations improve the expected utility of the advantaged population while harming that of the disadvantaged one. We next explore the introduction of a publicly-observable signal, such as a test score, as a potential intervention. Our main finding is that this natural intervention, intended to reduce the inequality between the populations' utilities, may actually exacerbate it in settings where observations and test scores are noisy.
{"title":"Access to Population-Level Signaling as a Source of Inequality","authors":"Nicole Immorlica, Katrina Ligett, Juba Ziani","doi":"10.1145/3287560.3287579","DOIUrl":"https://doi.org/10.1145/3287560.3287579","url":null,"abstract":"We identify and explore differential access to population-level signaling (also known as information design) as a source of unequal access to opportunity. A population-level signaler has potentially noisy observations of a binary type for each member of a population and, based on this, produces a signal about each member. A decision-maker infers types from signals and accepts those individuals whose type is high in expectation. We assume the signaler of the disadvantaged population reveals her observations to the decision-maker, whereas the signaler of the advantaged population forms signals strategically. We study the expected utility of the populations as measured by the fraction of accepted members, as well as the false positive rates (FPR) and false negative rates (FNR). We first show the intuitive results that for a fixed environment, the advantaged population has higher expected utility, higher FPR, and lower FNR, than the disadvantaged one (despite having identical population quality), and that more accurate observations improve the expected utility of the advantaged population while harming that of the disadvantaged one. We next explore the introduction of a publicly-observable signal, such as a test score, as a potential intervention. Our main finding is that this natural intervention, intended to reduce the inequality between the populations' utilities, may actually exacerbate it in settings where observations and test scores are noisy.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76388041","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}
David Madras, Elliot Creager, T. Pitassi, R. Zemel
How do we learn from biased data? Historical datasets often reflect historical prejudices; sensitive or protected attributes may affect the observed treatments and outcomes. Classification algorithms tasked with predicting outcomes accurately from these datasets tend to replicate these biases. We advocate a causal modeling approach to learning from biased data, exploring the relationship between fair classification and intervention. We propose a causal model in which the sensitive attribute confounds both the treatment and the outcome. Building on prior work in deep learning and generative modeling, we describe how to learn the parameters of this causal model from observational data alone, even in the presence of unobserved confounders. We show experimentally that fairness-aware causal modeling provides better estimates of the causal effects between the sensitive attribute, the treatment, and the outcome. We further present evidence that estimating these causal effects can help learn policies that are both more accurate and fair, when presented with a historically biased dataset.
{"title":"Fairness through Causal Awareness: Learning Causal Latent-Variable Models for Biased Data","authors":"David Madras, Elliot Creager, T. Pitassi, R. Zemel","doi":"10.1145/3287560.3287564","DOIUrl":"https://doi.org/10.1145/3287560.3287564","url":null,"abstract":"How do we learn from biased data? Historical datasets often reflect historical prejudices; sensitive or protected attributes may affect the observed treatments and outcomes. Classification algorithms tasked with predicting outcomes accurately from these datasets tend to replicate these biases. We advocate a causal modeling approach to learning from biased data, exploring the relationship between fair classification and intervention. We propose a causal model in which the sensitive attribute confounds both the treatment and the outcome. Building on prior work in deep learning and generative modeling, we describe how to learn the parameters of this causal model from observational data alone, even in the presence of unobserved confounders. We show experimentally that fairness-aware causal modeling provides better estimates of the causal effects between the sensitive attribute, the treatment, and the outcome. We further present evidence that estimating these causal effects can help learn policies that are both more accurate and fair, when presented with a historically biased dataset.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90832895","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}
Hadi Elzayn, S. Jabbari, Christopher Jung, Michael Kearns, S. Neel, Aaron Roth, Zachary Schutzman
Settings such as lending and policing can be modeled by a centralized agent allocating a scarce resource (e.g. loans or police officers) amongst several groups, in order to maximize some objective (e.g. loans given that are repaid, or criminals that are apprehended). Often in such problems fairness is also a concern. One natural notion of fairness, based on general principles of equality of opportunity, asks that conditional on an individual being a candidate for the resource in question, the probability of actually receiving it is approximately independent of the individual's group. For example, in lending this would mean that equally creditworthy individuals in different racial groups have roughly equal chances of receiving a loan. In policing it would mean that two individuals committing the same crime in different districts would have roughly equal chances of being arrested. In this paper, we formalize this general notion of fairness for allocation problems and investigate its algorithmic consequences. Our main technical results include an efficient learning algorithm that converges to an optimal fair allocation even when the allocator does not know the frequency of candidates (i.e. creditworthy individuals or criminals) in each group. This algorithm operates in a censored feedback model in which only the number of candidates who received the resource in a given allocation can be observed, rather than the true number of candidates in each group. This models the fact that we do not learn the creditworthiness of individuals we do not give loans to and do not learn about crimes committed if the police presence in a district is low. As an application of our framework and algorithm, we consider the predictive policing problem, in which the resource being allocated to each group is the number of police officers assigned to each district. The learning algorithm is trained on arrest data gathered from its own deployments on previous days, resulting in a potential feedback loop that our algorithm provably overcomes. In this case, the fairness constraint asks that the probability that an individual who has committed a crime is arrested should be independent of the district in which they live. We investigate the performance of our learning algorithm on the Philadelphia Crime Incidents dataset.
{"title":"Fair Algorithms for Learning in Allocation Problems","authors":"Hadi Elzayn, S. Jabbari, Christopher Jung, Michael Kearns, S. Neel, Aaron Roth, Zachary Schutzman","doi":"10.1145/3287560.3287571","DOIUrl":"https://doi.org/10.1145/3287560.3287571","url":null,"abstract":"Settings such as lending and policing can be modeled by a centralized agent allocating a scarce resource (e.g. loans or police officers) amongst several groups, in order to maximize some objective (e.g. loans given that are repaid, or criminals that are apprehended). Often in such problems fairness is also a concern. One natural notion of fairness, based on general principles of equality of opportunity, asks that conditional on an individual being a candidate for the resource in question, the probability of actually receiving it is approximately independent of the individual's group. For example, in lending this would mean that equally creditworthy individuals in different racial groups have roughly equal chances of receiving a loan. In policing it would mean that two individuals committing the same crime in different districts would have roughly equal chances of being arrested. In this paper, we formalize this general notion of fairness for allocation problems and investigate its algorithmic consequences. Our main technical results include an efficient learning algorithm that converges to an optimal fair allocation even when the allocator does not know the frequency of candidates (i.e. creditworthy individuals or criminals) in each group. This algorithm operates in a censored feedback model in which only the number of candidates who received the resource in a given allocation can be observed, rather than the true number of candidates in each group. This models the fact that we do not learn the creditworthiness of individuals we do not give loans to and do not learn about crimes committed if the police presence in a district is low. As an application of our framework and algorithm, we consider the predictive policing problem, in which the resource being allocated to each group is the number of police officers assigned to each district. The learning algorithm is trained on arrest data gathered from its own deployments on previous days, resulting in a potential feedback loop that our algorithm provably overcomes. In this case, the fairness constraint asks that the probability that an individual who has committed a crime is arrested should be independent of the district in which they live. We investigate the performance of our learning algorithm on the Philadelphia Crime Incidents dataset.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77314793","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}
Filipe Nunes Ribeiro, Koustuv Saha, Mahmoudreza Babaei, Lucas Henrique, Johnnatan Messias, Fabrício Benevenuto, Oana Goga, K. Gummadi, Elissa M. Redmiles
Targeted advertising is meant to improve the efficiency of matching advertisers to their customers. However, targeted advertising can also be abused by malicious advertisers to efficiently reach people susceptible to false stories, stoke grievances, and incite social conflict. Since targeted ads are not seen by non-targeted and non-vulnerable people, malicious ads are likely to go unreported and their effects undetected. This work examines a specific case of malicious advertising, exploring the extent to which political ads1 from the Russian Intelligence Research Agency (IRA) run prior to 2016 U.S. elections exploited Facebook's targeted advertising infrastructure to efficiently target ads on divisive or polarizing topics (e.g., immigration, race-based policing) at vulnerable sub-populations. In particular, we do the following: (a) We conduct U.S. census-representative surveys to characterize how users with different political ideologies report, approve, and perceive truth in the content of the IRA ads. Our surveys show that many ads are "divisive": they elicit very different reactions from people belonging to different socially salient groups. (b) We characterize how these divisive ads are targeted to sub-populations that feel particularly aggrieved by the status quo. Our findings support existing calls for greater transparency of content and targeting of political ads. (c) We particularly focus on how the Facebook ad API facilitates such targeting. We show how the enormous amount of personal data Facebook aggregates about users and makes available to advertisers enables such malicious targeting.
{"title":"On Microtargeting Socially Divisive Ads: A Case Study of Russia-Linked Ad Campaigns on Facebook","authors":"Filipe Nunes Ribeiro, Koustuv Saha, Mahmoudreza Babaei, Lucas Henrique, Johnnatan Messias, Fabrício Benevenuto, Oana Goga, K. Gummadi, Elissa M. Redmiles","doi":"10.1145/3287560.3287580","DOIUrl":"https://doi.org/10.1145/3287560.3287580","url":null,"abstract":"Targeted advertising is meant to improve the efficiency of matching advertisers to their customers. However, targeted advertising can also be abused by malicious advertisers to efficiently reach people susceptible to false stories, stoke grievances, and incite social conflict. Since targeted ads are not seen by non-targeted and non-vulnerable people, malicious ads are likely to go unreported and their effects undetected. This work examines a specific case of malicious advertising, exploring the extent to which political ads1 from the Russian Intelligence Research Agency (IRA) run prior to 2016 U.S. elections exploited Facebook's targeted advertising infrastructure to efficiently target ads on divisive or polarizing topics (e.g., immigration, race-based policing) at vulnerable sub-populations. In particular, we do the following: (a) We conduct U.S. census-representative surveys to characterize how users with different political ideologies report, approve, and perceive truth in the content of the IRA ads. Our surveys show that many ads are \"divisive\": they elicit very different reactions from people belonging to different socially salient groups. (b) We characterize how these divisive ads are targeted to sub-populations that feel particularly aggrieved by the status quo. Our findings support existing calls for greater transparency of content and targeting of political ads. (c) We particularly focus on how the Facebook ad API facilitates such targeting. We show how the enormous amount of personal data Facebook aggregates about users and makes available to advertisers enables such malicious targeting.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75456209","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 study a two-stage model, in which students are 1) admitted to college on the basis of an entrance exam which is a noisy signal about their qualifications (type), and then 2) those students who were admitted to college can be hired by an employer as a function of their college grades, which are an independently drawn noisy signal of their type. Students are drawn from one of two populations, which might have different type distributions. We assume that the employer at the end of the pipeline is rational, in the sense that it computes a posterior distribution on student type conditional on all information that it has available (college admissions, grades, and group membership), and makes a decision based on posterior expectation. We then study what kinds of fairness goals can be achieved by the college by setting its admissions rule and grading policy. For example, the college might have the goal of guaranteeing equal opportunity across populations: that the probability of passing through the pipeline and being hired by the employer should be independent of group membership, conditioned on type. Alternately, the college might have the goal of incentivizing the employer to have a group blind hiring rule. We show that both goals can be achieved when the college does not report grades. On the other hand, we show that under reasonable conditions, these goals are impossible to achieve even in isolation when the college uses an (even minimally) informative grading policy.
{"title":"Downstream Effects of Affirmative Action","authors":"Sampath Kannan, Aaron Roth, Juba Ziani","doi":"10.1145/3287560.3287578","DOIUrl":"https://doi.org/10.1145/3287560.3287578","url":null,"abstract":"We study a two-stage model, in which students are 1) admitted to college on the basis of an entrance exam which is a noisy signal about their qualifications (type), and then 2) those students who were admitted to college can be hired by an employer as a function of their college grades, which are an independently drawn noisy signal of their type. Students are drawn from one of two populations, which might have different type distributions. We assume that the employer at the end of the pipeline is rational, in the sense that it computes a posterior distribution on student type conditional on all information that it has available (college admissions, grades, and group membership), and makes a decision based on posterior expectation. We then study what kinds of fairness goals can be achieved by the college by setting its admissions rule and grading policy. For example, the college might have the goal of guaranteeing equal opportunity across populations: that the probability of passing through the pipeline and being hired by the employer should be independent of group membership, conditioned on type. Alternately, the college might have the goal of incentivizing the employer to have a group blind hiring rule. We show that both goals can be achieved when the college does not report grades. On the other hand, we show that under reasonable conditions, these goals are impossible to achieve even in isolation when the college uses an (even minimally) informative grading policy.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81562169","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}
Lily Hu, Nicole Immorlica, Jennifer Wortman Vaughan
When consequential decisions are informed by algorithmic input, individuals may feel compelled to alter their behavior in order to gain a system's approval. Models of agent responsiveness, termed "strategic manipulation," analyze the interaction between a learner and agents in a world where all agents are equally able to manipulate their features in an attempt to "trick" a published classifier. In cases of real world classification, however, an agent's ability to adapt to an algorithm is not simply a function of her personal interest in receiving a positive classification, but is bound up in a complex web of social factors that affect her ability to pursue certain action responses. In this paper, we adapt models of strategic manipulation to capture dynamics that may arise in a setting of social inequality wherein candidate groups face different costs to manipulation. We find that whenever one group's costs are higher than the other's, the learner's equilibrium strategy exhibits an inequality-reinforcing phenomenon wherein the learner erroneously admits some members of the advantaged group, while erroneously excluding some members of the disadvantaged group. We also consider the effects of interventions in which a learner subsidizes members of the disadvantaged group, lowering their costs in order to improve her own classification performance. Here we encounter a paradoxical result: there exist cases in which providing a subsidy improves only the learner's utility while actually making both candidate groups worse-off---even the group receiving the subsidy. Our results reveal the potentially adverse social ramifications of deploying tools that attempt to evaluate an individual's "quality" when agents' capacities to adaptively respond differ.
{"title":"The Disparate Effects of Strategic Manipulation","authors":"Lily Hu, Nicole Immorlica, Jennifer Wortman Vaughan","doi":"10.1145/3287560.3287597","DOIUrl":"https://doi.org/10.1145/3287560.3287597","url":null,"abstract":"When consequential decisions are informed by algorithmic input, individuals may feel compelled to alter their behavior in order to gain a system's approval. Models of agent responsiveness, termed \"strategic manipulation,\" analyze the interaction between a learner and agents in a world where all agents are equally able to manipulate their features in an attempt to \"trick\" a published classifier. In cases of real world classification, however, an agent's ability to adapt to an algorithm is not simply a function of her personal interest in receiving a positive classification, but is bound up in a complex web of social factors that affect her ability to pursue certain action responses. In this paper, we adapt models of strategic manipulation to capture dynamics that may arise in a setting of social inequality wherein candidate groups face different costs to manipulation. We find that whenever one group's costs are higher than the other's, the learner's equilibrium strategy exhibits an inequality-reinforcing phenomenon wherein the learner erroneously admits some members of the advantaged group, while erroneously excluding some members of the disadvantaged group. We also consider the effects of interventions in which a learner subsidizes members of the disadvantaged group, lowering their costs in order to improve her own classification performance. Here we encounter a paradoxical result: there exist cases in which providing a subsidy improves only the learner's utility while actually making both candidate groups worse-off---even the group receiving the subsidy. Our results reveal the potentially adverse social ramifications of deploying tools that attempt to evaluate an individual's \"quality\" when agents' capacities to adaptively respond differ.","PeriodicalId":20573,"journal":{"name":"Proceedings of the Conference on Fairness, Accountability, and Transparency","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80660983","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：