Pith. sign in

REVIEW 3 major objections 5 minor 2 cited by

Data Management for Causal Algorithmic Fairness

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The concept of fairness requires causal reasoning, and associational definitions can mislead.

desk verdict A clear, well-written position paper that usefully frames fairness as a data management problem, but its central claim that fairness requires causal reasoning rests on causal assumptions that the paper's own examples do not justify. read the letter →

arxiv 1908.07924 v3 pith:YOMJLXLB submitted 2019-08-20 cs.DB cs.LG

Babak Salimi , Bill Howe , Dan Suciu This is my paper · ORCID
classification cs.DBcs.LG
keywords causalfairnessalgorithmicinterventionaljustifiableconditionalindependenceconstraintsdatabaserepairinferencedatamanagement
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that algorithmic fairness is fundamentally a causal question: whether a protected attribute such as gender or race causes an outcome cannot be settled by correlation, because associations can arise from mediating variables and confounding. It distinguishes associational fairness criteria (demographic parity, equalized odds, predictive parity) from causal criteria (counterfactual, proxy, path-specific, interventional fairness) and claims the associational ones can produce incorrect or misleading verdicts. The central technical claim is that K-fairness and justifiable fairness, defined by intervening on the protected attribute and on context variables, correctly classify two college-admission examples where associational measures fail. If right, fairness assessment and repair should happen in the data pipeline, by enforcing conditional-independence constraints, not just by post-processing a model.

What carries the argument

The central object is the causal DAG together with Pearl's do-operator, used to define K-fairness: an algorithm is K-fair if $Pr(O=o | do(S=0), do(K=k)) = Pr(O=o | do(S=1), do(K=k))$ for every context $k$; justifiable fairness requires this for every superset $K$ of the admissible variable set $A$. The companion machinery is the encoding of fairness as conditional-independence integrity constraints, where a missing arrow in the DAG corresponds to a conditional-independence statement, which turns fairness enforcement into a database repair problem. The CAPUCHIN system repairs training data by enforcing the conditional independence of the training label $Y$ and the inadmissible variables $I$ given the admissible variables $A$, reducing the task to a minimal repair under multivalued dependencies solved with MaxSAT and matrix factorization.

What would settle it

Run justifiable fairness and CAPUCHIN on a synthetic admissions dataset built from a known true graph that includes an unmeasured confounder of the admissible department variable and the outcome, but feed the system only the incomplete graph without that confounder; if the repaired data still trains a classifier whose outcomes differ under $do(G=1)$ versus $do(G=0)$ once the confounder is revealed, the framework's guarantee fails.

Watch

Extended reading notes

Core claim

The paper's core claim is that the concept of fairness requires causal reasoning, and that popular associational definitions in ML can produce incorrect or misleading results. In support, it presents two college-admission scenarios: College I, where each department admits by hobby and the college-wide admission rate is identical for men and women, so demographic parity and proxy fairness call it fair even though the process discriminates; College II, where conditioning on department falsely suggests discrimination but the only path from gender to outcome goes through the admissible variable department, so the process is fair. K-fairness, defined by equality of outcome probabilities under interventions $do(S=s)$ and $do(K=k)$, and its practical refinement justifiable fairness, based on a user-supplied partition into admissible and inadmissible variables, identify College I as discriminatory and College II as fair. The paper further claims that fairness can be treated as a set of conditional-independence integrity constraints, enabling data-management techniques such as query rewriting, database repair, weak supervision, and provenance to detect and enforce causal fairness.

Load-bearing premise

The load-bearing premise is that the user's picture of cause and effect, and the user's list of which factors it is permissible for the protected attribute to influence, are both correct; if a hidden cause is missing or the list is disputed, the fairness verdicts and the repair guarantee do not hold.

Editorial extensions

If this is right

  • Fairness audits based only on demographic parity, equalized odds, or predictive parity can certify a discriminatory process as fair (College I) or condemn a fair process as discriminatory (College II).
  • Causal fairness criteria require users to specify which variables are admissible pathways for the protected attribute, so the outcome of any fairness audit depends on that normative choice.
  • A sufficient condition for a classifier trained on repaired data to be justifiably fair is that the training distribution satisfies the conditional independence of $Y$ and $I$ given $A$. The CAPUCHIN repair methods reduce bias with smaller accuracy loss than dropping the sensitive and inadmissible variables entirely.
  • Filtering and conditioning can introduce bias even when raw data are unbiased, because conditional independencies are not closed under conditioning, so fairness constraints must be enforced through the whole transformation pipeline.
  • Detecting discrimination requires causal hypothesis tests on the effect of protected-attribute membership, and the HYPDB system shows how ordinary SQL group-by queries can be rewritten into such causal queries.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • An implication left implicit is that every fairness claim should be expected to come with its causal graph and its admissibility partition, because two reasonable users could reach opposite verdicts on the same data from different partitions.
  • A testable extension would be to apply the College I and College II benchmark to real admissions or hiring data with known decision rules, measuring how often associational metrics and K-fairness diverge.
  • The conditional-independence-as-integrity-constraint view suggests fairness repair could be embedded in standard ETL pipelines as a declarative constraint, and raises the possibility of learning admissibility sets from data, though that would change the normative character of the definition.
  • If the user-supplied graph is incomplete, justifiable fairness may be worse than associational metrics because it gives false confidence; hidden confounders between an admissible variable and the outcome are the natural failure mode to test.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This short position paper argues that algorithmic fairness cannot be captured by associational criteria and that 'the concept of fairness requires causal reasoning.' It distinguishes associational definitions (demographic parity, conditional statistical parity, equalized odds, predictive parity) from causal definitions (counterfactual, proxy, path-specific, and interventional fairness), introduces the notion of justifiable fairness based on a user-supplied partition of variables into admissible and inadmissible, and illustrates the distinction with two synthetic college-admission examples. The second half reviews data management opportunities: treating causal fairness constraints as conditional-independence integrity constraints, query rewriting for detecting discrimination (HYPDB), database repair for enforcing fairness (CAPUCHIN), fairness-aware weak supervision, and provenance for explaining bias.

Significance. If the central thesis is accepted, the paper provides a useful agenda: fairness constraints become conditional-independence integrity constraints that can be enforced by data-repair techniques, and standard OLAP-style analyses are argued to be misleading without causal adjustment. The manuscript is technically careful in its use of d-separation and the do-operator, and the College I computations for Eq. (4) check out. Its main contribution at this stage is conceptual and programmatic rather than a new formal result; the value depends on the user's ability to specify a correct causal DAG and admissible set, a limitation the paper acknowledges only in passing.

major comments (3)
  1. [Section 2.2, Example 2 (Fig. 2)] The assertion that justifiable fairness 'correctly identifies College II as fair' is underdetermined by the observed data. The same admission table is compatible with a causal DAG that includes an edge from gender G to qualification Q, with Q inadmissible, in which case K={D}-fairness fails because the distribution of Q would change under do(G=1) versus do(G=0). The example therefore shows only that a particular causal model yields a particular verdict; it does not by itself demonstrate that causal definitions recover hidden structure. Please qualify the claim as conditional on the assumed DAG and admissible set, and discuss sensitivity to DAG misspecification.
  2. [Section 3.1] Missing arrows are treated as assumptions of no causal effect, but the manuscript does not discuss how the causal DAG is learned or validated. Since every subsequent proposal, including the CAPUCHIN repair, inherits this assumption, the paper should state this limitation explicitly and note that an incorrect DAG can lead to enforcing the wrong conditional-independence constraints. Without such a caveat, the proposed data-management pipeline appears to deliver fairness guarantees that actually depend on an unstated modeling choice.
  3. [Section 3.3] The statement that models trained on CAPUCHIN-repaired data 'can be shown to be justifiably fair even on unseen test data under some mild assumptions' is not self-contained; the assumptions are not stated in this manuscript, so the reader cannot assess the scope of the guarantee. At minimum, name the assumptions or refer precisely to the theorem in [43] and state its preconditions, including any stability or no-confounder requirements.
minor comments (5)
  1. [Section 2.2] The phrase 'casual DAG' in the second paragraph should be 'causal DAG'.
  2. [Example 2] The displayed formula contains an extra closing parenthesis after 'Q = q'; it should read Pr(O = 1|G = i, D = d, Q = q) Pr(Q = q|G = i).
  3. [Section 3.2.2] The text refers to 'Adult data described in Ex 1', but Example 1 is the Berkeley admissions case; the reference should be to Example 3.
  4. [Figure 4 caption] The caption uses 'Bias(ROD)' without defining ROD in the text; please define 'Ratio of Observational Discrimination' at first use.
  5. [References] Reference [20] lacks a URL or a more complete citation for the UCI repository.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the paper's causal-fairness claims are definitional and explicitly assumption-dependent, and its core critique is anchored by an external impossibility result.

full rationale

The paper is a position/review article, not a fitted empirical derivation. Its central claim that 'the concept of fairness requires causal reasoning' is supported by the Berkeley admissions example, the COMPAS debate, and Chouldechova's external impossibility theorem (Section 2.1), none of which are supplied by the authors' own fitted values. The interventional and justifiable fairness definitions in Eq. 4 are stated directly in Section 2.2 and are explicitly conditional on a user-supplied admissible/inadmissible partition ('This partitioning expresses fairness social norms and values and comes from the users'), so the College I/II classifications are illustrations of the definition under stated causal assumptions, not hidden predictions derived from fitted inputs. The CAPUCHIN, HYPDB, and ROD material is heavily self-cited from the authors' prior work, and the guarantee in Section 3.3 that repaired data yields justifiable fairness 'under some mild assumptions' is imported without stating those assumptions; however, no specific equation or prediction in this paper reduces by construction to a fitted parameter or to a self-citation chain. The underdetermination of the causal DAG noted by critics is a real correctness/assumption risk, but it is not a circularity: the paper openly treats the DAG and admissibility set as inputs. We therefore find no circular step, but assign score 2 to reflect the load-bearing role of the authors' own unpublished-in-this-paper CAPUCHIN guarantee and the heavy reliance on self-citations for the operative fairness definitions.

Assumptions & free parameters 0 free parameters · 4 assumptions · 2 invented entities

The central claim rests on several domain assumptions: correctness of the causal DAG, absence of hidden confounders, and a user-supplied admissibility partition. No numerical free parameters are fitted in this paper; the CAPUCHIN experiment parameters are described in cited prior work. The main invented constructs, justifiable fairness and the ROD metric, come from the authors' prior papers and have no independent falsifiable handle in this manuscript.

assumptions (4)
  • domain assumption The correct causal DAG over the variable set V is known, or at least the missing arrows are known to represent absence of causal effect.
    Invoked throughout Sections 2.2 and 3.1: 'It is important to know the causal DAG since the probability distribution is insufficient to compute the do operator.' The entire framework uses d-separation and do-operator on a presumed graph.
  • domain assumption The partition of variables into admissible A and inadmissible I is supplied by users and correctly encodes fairness norms.
    Section 2.2 states 'This partitioning expresses fairness social norms and values and comes from the users.' Justifiable fairness and CAPUCHIN repairs depend on this partition being correct.
  • domain assumption No unmeasured confounding or hidden common causes invalidate the conditional independence statements derived from missing edges.
    Section 3.1 equates missing edges with conditional independence. If unobserved confounders exist, the do-operator computations and the repairs are not valid. The paper never tests this assumption.
  • ad hoc to paper Under 'some mild assumptions' from [43], models trained on CAPUCHIN-repaired data remain justifiably fair on unseen test data.
    Section 3.3 imports this guarantee from the authors' prior SIGMOD paper without stating the assumptions or proof in this manuscript.
invented entities (2)
  • Interventional fairness / K-fairness / justifiable fairness (Eq. 4)
    purpose: Define group-level causal fairness by requiring Pr(O | do(S=0), do(K=k)) = Pr(O | do(S=1), do(K=k)) for supersets K of the admissible variables, avoiding individual-level counterfactuals.
    Presented in Section 2.2 as the definition that resolves the College examples. It is a normative definition from the authors' prior work [43], not an observable quantity with a falsifiable prediction outside the paper.
  • Ratio of Observational Discrimination (ROD)
    purpose: Quantifies the effect of the protected attribute on the classifier outcome while controlling for admissible variables; used as the bias metric in Fig. 4.
    This metric comes from the authors' prior work [42,43] and is not independently validated outside that work. Using it to evaluate the authors' own repairs is a mild self-referential component of the evaluation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Data Management for Causal Algorithmic Fairness." pith.science (2026). https://pith.science/paper/YOMJLXLB

@misc{pith2026190807924,
  author       = {Pith},
  title        = {Pith review of: Data Management for Causal Algorithmic Fairness},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YOMJLXLB}},
  note         = {Machine review of arXiv:1908.07924}
}
read the original abstract

Fairness is increasingly recognized as a critical component of machine learning systems. However, it is the underlying data on which these systems are trained that often reflects discrimination, suggesting a data management problem. In this paper, we first make a distinction between associational and causal definitions of fairness in the literature and argue that the concept of fairness requires causal reasoning. We then review existing works and identify future opportunities for applying data management techniques to causal algorithmic fairness.

Figures

Figures reproduced from arXiv: 1908.07924 by the authors.

Figure 1
Figure 1. Common associational definitions of fairness. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Admission process representation in two colleges where associational fairness fail (see Ex.2). [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. (a) HYPDB’s report on the effect of gender on income (cf. Ex. 1). (b) A compact causal DAG with O = income, G = gender, M = marital status, C = age and nationality, E = education and W = work class, occupation and hours per week (cf. Ex. 3). 3.2 Query Rewriting In data management, query rewriting refers to a set of techniques to automatically modify one query into another that satisfies certain desired properties. T… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Performance of CAPUCHIN on Adult data. of tuples) to obtain another training database D0 that satisfies (Y ⊥⊥I|A). Specifically, they reduced the problem to a minimal repair problem w.r.t. an MVD and developed a set of techniques, including reduction to the MaxSAT and …

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Fair-Count-Min: Frequency Estimation under Equal Group-wise Approximation Factor

    cs.DS 2025-05 reject novelty 6.0 of 10

    Fair-Count-Min partitions Count-Min columns among groups, allocating columns by group size for one hash row and by a binomial-minimum equation for multiple rows, aiming to equalize expected approximation factors.

  2. The Fair Game: Auditing & Debiasing AI Algorithms Over Time

    cs.AI 2025-08 unverdicted novelty 4.0 of 10

    Proposes 'Fair Game', a reinforcement-learning loop in which an auditor's bias criteria, updatable over time, steer a debiasing agent that adapts an ML model's predictions.

Reference graph

Works this paper leans on

48 extracted references · 40 canonical work pages · cited by 2 Pith papers

  1. [43]

    Interventional fairness: Causal database repair for algorithmic fairness

    Babak Salimi, Luke Rodriguez, Bill Howe, and Dan Suciu. Interventional fairness: Causal database repair for algorithmic fairness. In Proceedings of the 2019 International Conference on Management of Data, pages 793–810. ACM, 2019

    work page 2019

  2. [1]

    Foundations of Databases

    Serge Abiteboul, Richard Hull, and Victor Vianu. Foundations of Databases. Addison-Wesley, 1995

    1995

  3. [2]

    Approximated summarization of data provenance

    Eleanor Ainy, Pierre Bourhis, Susan B Davidson, Daniel Deutch, and Tova Milo. Approximated summarization of data provenance. In Proceedings of the 24th ACM International on Conference on Information and Knowledge Management, pages 483–492. ACM, 2015

    work page 2015

  4. [3]

    Constraint-based explanation and repair of filter-based transformations

    Dolan Antenucci and Michael Cafarella. Constraint-based explanation and repair of filter-based transformations. Proceedings of the VLDB Endowment, 11(9):947–960, 2018

    work page 2018

  5. [4]

    Identifiability of path-specific effects

    Chen Avin, Ilya Shpitser, and Judea Pearl. Identifiability of path-specific effects. 2005

    work page 2005

  6. [5]

    Causes for query answers from databases: Datalog abduction, view-updates, and integrity constraints

    Leopoldo Bertossi and Babak Salimi. Causes for query answers from databases: Datalog abduction, view-updates, and integrity constraints. International Journal of Approximate Reasoning, 90:226–252, 2017

    work page 2017

  7. [6]

    Bertossi

    Leopoldo E. Bertossi. Database Repairing and Consistent Query Answering . Synthesis Lectures on Data Manage- ment. Morgan & Claypool Publishers, 2011. 10

    work page 2011

  8. [7]

    Building classifiers with independency constraints

    Toon Calders, Faisal Kamiran, and Mykola Pechenizkiy. Building classifiers with independency constraints. In Data mining workshops, 2009. ICDMW’09. IEEE international conference on, pages 13–18. IEEE, 2009

    work page 2009

Show all 48 references
  1. [8]

    Why not? In Proceedings of the 2009 ACM SIGMOD International Conference on Management of data, pages 523–534

    Adriane Chapman and HV Jagadish. Why not? In Proceedings of the 2009 ACM SIGMOD International Conference on Management of data, pages 523–534. ACM, 2009

    2009

  2. [9]

    Fair prediction with disparate impact: A study of bias in recidivism prediction instruments

    Alexandra Chouldechova. Fair prediction with disparate impact: A study of bias in recidivism prediction instruments. Big data, 5(2):153–163, 2017

    2017

  3. [10]

    Algorithmic decision making and the cost of fairness

    Sam Corbett-Davies, Emma Pierson, Avi Feller, Sharad Goel, and Aziz Huq. Algorithmic decision making and the cost of fairness. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 797–806. ACM, 2017

    2017

  4. [11]

    Bias detectives: the researchers striving to make algorithms fair

    Rachel Courtland. Bias detectives: the researchers striving to make algorithms fair. Nature, 558, 2018

    2018

  5. [12]

    Fairness through awareness

    Cynthia Dwork, Moritz Hardt, Toniann Pitassi, Omer Reingold, and Richard Zemel. Fairness through awareness. In Proceedings of the 3rd innovations in theoretical computer science conference, pages 214–226. ACM, 2012

    2012

  6. [13]

    Fairness testing: testing software for discrimination

    Sainyam Galhotra, Yuriy Brun, and Alexandra Meliou. Fairness testing: testing software for discrimination. In Proceedings of the 2017 11th Joint Meeting on Foundations of Software Engineering, pages 498–510. ACM, 2017

    2017

  7. [14]

    Data provenance: A categorization of existing approaches

    Boris Glavic and Klaus Dittrich. Data provenance: A categorization of existing approaches. Datenbanksysteme in Business, Technologie und Web (BTW 2007)–12. Fachtagung des GI-Fachbereichs” Datenbanken und Information- ssysteme”(DBIS), 2007

    2007

  8. [15]

    Equality of opportunity in supervised learning

    Moritz Hardt, Eric Price, Nati Srebro, et al. Equality of opportunity in supervised learning. In Advances in neural information processing systems, pages 3315–3323, 2016

    2016

  9. [16]

    Avoiding discrimination through causal reasoning

    Niki Kilbertus, Mateo Rojas Carulla, Giambattista Parascandolo, Moritz Hardt, Dominik Janzing, and Bernhard Sch¨olkopf. Avoiding discrimination through causal reasoning. In Advances in Neural Information Processing Sys- tems, pages 656–666, 2017

    2017

  10. [17]

    Counterfactual fairness

    Matt J Kusner, Joshua Loftus, Chris Russell, and Ricardo Silva. Counterfactual fairness. In Advances in Neural Information Processing Systems, pages 4069–4079, 2017

    2017

  11. [18]

    How we analyzed the compas recidivism algorithm

    Jeff Larson, Surya Mattu, Lauren Kirchner, and Julia Angwin. How we analyzed the compas recidivism algorithm. ProPublica (5 2016), 9, 2016

    2016

  12. [19]

    Levy, Alberto O

    Alon Y . Levy, Alberto O. Mendelzon, Yehoshua Sagiv, and Divesh Srivastava. Answering queries using views. In Proceedings of the Fourteenth ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems, May 22-25, 1995, San Jose, California, USA, pages 95–104, 1995

    1995

  13. [20]

    M. Lichman. Uci machine learning repository, 2013

    2013

  14. [21]

    Computing optimal repairs for functional dependencies

    Ester Livshits, Benny Kimelfeld, and Sudeepa Roy. Computing optimal repairs for functional dependencies. In Proceedings of the 37th ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems, Houston, TX, USA, June 10-15, 2018, pages 225–237, 2018

    2018

  15. [22]

    Causal reasoning for algorithmic fairness

    Joshua R Loftus, Chris Russell, Matt J Kusner, and Ricardo Silva. Causal reasoning for algorithmic fairness. arXiv preprint arXiv:1805.05859, 2018

    2018 arXiv

  16. [23]

    The complexity of causality and responsibility for query answers and non-answers

    Alexandra Meliou, Wolfgang Gatterbauer, Katherine F Moore, and Dan Suciu. The complexity of causality and responsibility for query answers and non-answers. Proceedings of the VLDB Endowment, 4(1):34–45, 2010

    2010

  17. [24]

    Distant supervision for relation extraction without labeled data

    Mike Mintz, Steven Bills, Rion Snow, and Dan Jurafsky. Distant supervision for relation extraction without labeled data. In Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of th...

    2009

  18. [25]

    Fair inference on outcomes

    Razieh Nabi and Ilya Shpitser. Fair inference on outcomes. In Proceedings of the... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence, volume 2018, page 1931. NIH Public Access, 2018

    2018

  19. [26]

    Causality: models, reasoning, and inference

    Judea Pearl. Causality: models, reasoning, and inference. Econometric Theory, 19(675-685):46, 2003. 11

    2003

  20. [27]

    Causality

    Judea Pearl. Causality. Cambridge university press, 2009

    2009

  21. [28]

    Causal inference in statistics: An overview

    Judea Pearl et al. Causal inference in statistics: An overview. Statistics Surveys, 3:96–146, 2009

    2009

  22. [29]

    A chase too far? In Proceedings of the 2000 ACM SIGMOD International Conference on Management of Data, May 16-18, 2000, Dallas, Texas, USA., pages 273–284, 2000

    Lucian Popa, Alin Deutsch, Arnaud Sahuguet, and Val Tannen. A chase too far? In Proceedings of the 2000 ACM SIGMOD International Conference on Management of Data, May 16-18, 2000, Dallas, Texas, USA., pages 273–284, 2000

    2000

  23. [30]

    Snorkel: Rapid training data creation with weak supervision

    Alexander Ratner, Stephen H Bach, Henry Ehrenberg, Jason Fries, Sen Wu, and Christopher R ´e. Snorkel: Rapid training data creation with weak supervision. Proceedings of the VLDB Endowment, 11(3):269–282, 2017

    2017

  24. [31]

    Data programming: Creating large training sets, quickly

    Alexander J Ratner, Christopher M De Sa, Sen Wu, Daniel Selsam, and Christopher R´e. Data programming: Creating large training sets, quickly. In Advances in neural information processing systems, pages 3567–3575, 2016

    2016

  25. [32]

    Learning from crowds

    Vikas C Raykar, Shipeng Yu, Linda H Zhao, Gerardo Hermosillo Valadez, Charles Florin, Luca Bogoni, and Linda Moy. Learning from crowds. Journal of Machine Learning Research, 11(Apr):1297–1322, 2010

    2010

  26. [33]

    A formal approach to finding explanations for database queries

    Sudeepa Roy and Dan Suciu. A formal approach to finding explanations for database queries. In Proceedings of the 2014 ACM SIGMOD international conference on Management of data, pages 1579–1590. ACM, 2014

    2014

  27. [34]

    The Use of Matched Sampling and Regression Adjustment in Observational Studies

    Donald B Rubin. The Use of Matched Sampling and Regression Adjustment in Observational Studies. Ph.D. Thesis, Department of Statistics, Harvard University, Cambridge, MA, 1970

    1970

  28. [35]

    Statistics and causal inference: Comment: Which ifs have causal answers

    Donald B Rubin. Statistics and causal inference: Comment: Which ifs have causal answers. Journal of the American Statistical Association, 81(396):961–962, 1986

    1986

  29. [36]

    Comment: The design and analysis of gold standard randomized experiments

    Donald B Rubin. Comment: The design and analysis of gold standard randomized experiments. Journal of the American Statistical Association, 103(484):1350–1353, 2008

    2008

  30. [37]

    When worlds collide: integrating different counter- factual assumptions in fairness

    Chris Russell, Matt J Kusner, Joshua Loftus, and Ricardo Silva. When worlds collide: integrating different counter- factual assumptions in fairness. In Advances in Neural Information Processing Systems, pages 6414–6423, 2017

    2017

  31. [38]

    Bertossi

    Babak Salimi and Leopoldo E. Bertossi. From causes for database queries to repairs and model-based diagnosis and back. In ICDT, pages 342–362, 2015

    2015

  32. [39]

    Quantifying causal effects on query answering in databases

    Babak Salimi, Leopoldo E Bertossi, Dan Suciu, and Guy Van den Broeck. Quantifying causal effects on query answering in databases. In TaPP, 2016

    2016

  33. [40]

    Hypdb: a demonstration of detecting, explain- ing and resolving bias in olap queries

    Babak Salimi, Corey Cole, Peter Li, Johannes Gehrke, and Dan Suciu. Hypdb: a demonstration of detecting, explain- ing and resolving bias in olap queries. Proceedings of the VLDB Endowment, 11(12):2062–2065, 2018

    2018

  34. [41]

    Bias in olap queries: Detection, explanation, and removal

    Babak Salimi, Johannes Gehrke, and Dan Suciu. Bias in olap queries: Detection, explanation, and removal. In Proceedings of the 2018 International Conference on Management of Data, pages 1021–1035. ACM, 2018

    2018

  35. [42]

    Capuchin: Causal database repair for algorithmic fairness

    Babak Salimi, Luke Rodriguez, Bill Howe, and Dan Suciu. Capuchin: Causal database repair for algorithmic fairness. arXiv preprint arXiv:1902.08283, 2019

    1902 arXiv

  36. [44]

    The problem of infra-marginality in outcome tests for discrimination

    Camelia Simoiu, Sam Corbett-Davies, Sharad Goel, et al. The problem of infra-marginality in outcome tests for discrimination. The Annals of Applied Statistics, 11(3):1193–1216, 2017

    2017

  37. [45]

    Fairness definitions explained

    Sahil Verma and Julia Rubin. Fairness definitions explained. In 2018 IEEE/ACM International Workshop on Software Fairness (FairWare), pages 1–7. IEEE, 2018

    2018

  38. [46]

    Provenance analysis for missing answers and integrity repairs

    Jane Xu, Waley Zhang, Abdussalam Alawini, and Val Tannen. Provenance analysis for missing answers and integrity repairs. IEEE Data Eng. Bull., 41(1):39–50, 2018

    2018

  39. [47]

    Fairness beyond dis- parate treatment & disparate impact: Learning classification without disparate mistreatment

    Muhammad Bilal Zafar, Isabel Valera, Manuel Gomez Rodriguez, and Krishna P Gummadi. Fairness beyond dis- parate treatment & disparate impact: Learning classification without disparate mistreatment. In Proceedings of the 26th International Conference on World Wide Web, pages 117...

    2017

  40. [48]

    A survey on measuring indirect discrimination in machine learning

    Indre Zliobaite. A survey on measuring indirect discrimination in machine learning. arXiv preprint arXiv:1511.00148, 2015. 12

    2015 arXiv

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.