pith. sign in

arxiv: 2607.00220 · v1 · pith:VI4WBH5Onew · submitted 2026-06-30 · 🧮 math.CT · cs.AI· cs.CY

A Category Theory Account of AI Identity

Andrea Ferrario This is my paper

Pith reviewed 2026-07-02 00:24 UTC · model grok-4.3

classification 🧮 math.CT cs.AIcs.CY
keywords AI identitycategory theorytrustworthiness profilediachronic identitysynchronic identityresponsible AIlifecycle pathsnatural transformations
0
0 comments X

The pith

Category theory replaces a single AI identity relation with a structured hierarchy of diachronic and synchronic criteria based on trustworthiness-preserving transformations.

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

The paper develops a category-theoretic formalization of AI identity to address when a system remains the same over time or across deployments. It begins from an earlier approach that ties identity to equality of trustworthiness levels but leaves the structure of states, transformations, and persistence implicit. An AI system type is defined by a datum of techno-function, trustworthiness profile, and level function; states connect via admissible lifecycle paths that preserve levels and form a reachability category after quotienting. Histories are modeled as temporally admissible functors and compared by time-synchronous natural transformations. This setup produces weak and strong categorical readings of the identity criteria and supplies preconditions for transferring responsible-AI claims without treating categorical identity as sufficient on its own.

Core claim

An AI system type is specified by a datum consisting of a techno-function, a trustworthiness profile, and a trustworthiness-level function. Profile-relative states are connected by admissible lifecycle paths, which are restricted to trustworthiness-level-preserving transformations and quotiented to obtain a reachability category. Temporally admissible functors represent AI system histories, while time-synchronous natural transformations compare realized histories. The formalization yields two categorical interpretations of the earlier AI identity criteria. A weak interpretation recovers identity as equality of trustworthiness level. A strong interpretation requires mutual trustworthiness-pre

What carries the argument

The reachability category, formed by quotienting admissible lifecycle paths that preserve trustworthiness levels, together with temporally admissible functors and time-synchronous natural transformations, which organize states, transformations, and histories to interpret identity criteria.

If this is right

  • Category theory replaces a single AI identity relation with a structured hierarchy of diachronic and synchronic criteria.
  • The framework identifies identity-related preconditions for transferring responsible-AI claims, evidence, and governance procedures across versions or deployments.
  • Categorical identity is not treated as sufficient by itself for such transfer.
  • A weak interpretation recovers identity as equality of trustworthiness level.
  • A strong interpretation requires mutual trustworthiness-preserving reachability expressed through isomorphism.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same categorical organization of states and transformations could be tested on non-AI technical systems that evolve through updates, such as control software in vehicles.
  • Governance bodies could check concrete deployments against the isomorphism conditions before moving oversight procedures between versions.
  • The approach suggests developing verification procedures that computationally inspect whether two AI histories are related by a time-synchronous natural transformation.

Load-bearing premise

The structure of AI system states, the admissible transformations between them, and the temporal organization of persistence can be adequately captured by the introduced datum, admissible lifecycle paths, reachability category, temporally admissible functors, and time-synchronous natural transformations.

What would settle it

An AI deployment in which trustworthiness levels match across versions yet the reachability category yields no state isomorphism, or in which responsibility transfers succeed despite failing the strong categorical conditions, would challenge the interpretations.

Figures

Figures reproduced from arXiv: 2607.00220 by Andrea Ferrario.

Figure 1
Figure 1. Figure 1: Left: two normalized trustworthiness dimensions, here accuracy and robustness, evolve over time under [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Two distinct lifecycle paths connecting the same abstract states [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: An AI system history functor A♢ : Tt0 → Sys♢. The states A♢(t) and A♢(t ′ ) lie on the same trustworthiness level k, represented by the horizontal line τ♢ = k. The lower connected path f A t,t′ is an A♢-time-ordered representative of the morphism At,t′ . The upper connected path illustrates another trustworthiness-level-preserving representative in the same equivalence class. (Note that this representative… view at source ↗
Figure 4
Figure 4. Figure 4: The upper part of the figure represents a natural transformation [PITH_FULL_IMAGE:figures/full_fig_p017_4.png] view at source ↗
read the original abstract

Artificial intelligence (AI) systems are routinely modified after deployment through retraining and changes in their environments. These transformations raise a metaphysical question: under what conditions does an AI system remain the same system over time or across deployments? Earlier work formulates synchronic and diachronic identity propositionally, by relating identity within a fixed AI system type to equality of trustworthiness levels. Such criteria specify when identity statements are true, but leave implicit the structure of the states compared, the transformations connecting them, and the temporal organization of persistence. We develop a category-theoretic formalization of AI identity. An AI system type is specified by a datum consisting of a techno-function, a trustworthiness profile, and a trustworthiness-level function. Profile-relative states are connected by admissible lifecycle paths, which are restricted to trustworthiness-level-preserving transformations and quotiented to obtain a reachability category. Temporally admissible functors represent AI system histories, while time-synchronous natural transformations compare realized histories. The formalization yields two categorical interpretations of the earlier AI identity criteria. A weak interpretation recovers identity as equality of trustworthiness level. A strong interpretation requires mutual trustworthiness-preserving reachability, expressed through state isomorphism or natural isomorphism of realized histories. Category theory therefore replaces a single AI identity relation with a structured hierarchy of diachronic and synchronic criteria. The resulting framework identifies identity-related preconditions for transferring responsible-AI claims, evidence, and governance procedures across versions or deployments, without treating categorical identity as sufficient by itself for such transfer.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper develops a category-theoretic formalization of AI identity. An AI system type is given by a datum consisting of a techno-function, trustworthiness profile, and trustworthiness-level function. Admissible lifecycle paths (restricted to trustworthiness-level-preserving transformations) are quotiented to form a reachability category. AI system histories are represented as temporally admissible functors, and comparisons between histories use time-synchronous natural transformations. This yields a weak interpretation of identity (equality of trustworthiness level) and a strong interpretation (mutual trustworthiness-preserving reachability via state or history isomorphism), replacing a single identity relation with a hierarchy of diachronic and synchronic criteria that identifies preconditions for transferring responsible-AI claims across versions or deployments.

Significance. If the constructions are adequate, the framework supplies a structured, parameter-free categorical hierarchy that refines earlier propositional criteria for AI identity. The explicit use of reachability categories, functors, and natural transformations to organize persistence and admissible change is a clear strength, as is the explicit disclaimer that categorical identity is not treated as sufficient for governance transfer. This could provide a formal basis for reasoning about continuity in AI systems without reducing to fitted parameters.

major comments (2)
  1. [Abstract] Abstract, paragraph describing the formalization: the central claim that the framework 'identifies identity-related preconditions for transferring responsible-AI claims' rests on the unverified assumption that the datum, reachability category, temporally admissible functors, and time-synchronous natural transformations faithfully encode governance-relevant notions of persistence and admissible change. No concrete construction or example is supplied showing how a specific AI system's states and transformations (e.g., non-deterministic retraining) map onto these objects or how the quotient by trustworthiness-level-preserving paths aligns with continuity relevant to evidence or procedure transfer.
  2. [Abstract] Abstract, strong interpretation paragraph: the requirement of 'mutual trustworthiness-preserving reachability, expressed through state isomorphism or natural isomorphism of realized histories' is presented as delivering the hierarchy, but without an explicit example of an admissible lifecycle path or a time-synchronous natural transformation, it is impossible to verify whether the isomorphism condition captures or abstracts away environment-dependent trustworthiness shifts that would block claim transfer in practice.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and for identifying areas where the connection between the categorical constructions and governance considerations could be made more explicit. We address each major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract, paragraph describing the formalization: the central claim that the framework 'identifies identity-related preconditions for transferring responsible-AI claims' rests on the unverified assumption that the datum, reachability category, temporally admissible functors, and time-synchronous natural transformations faithfully encode governance-relevant notions of persistence and admissible change. No concrete construction or example is supplied showing how a specific AI system's states and transformations (e.g., non-deterministic retraining) map onto these objects or how the quotient by trustworthiness-level-preserving paths aligns with continuity relevant to evidence or procedure transfer.

    Authors: The abstract condenses the framework whose definitions appear in the body of the manuscript (datum in Section 2, reachability category via quotient by level-preserving paths in Section 3, temporally admissible functors and time-synchronous natural transformations in Section 4). The claim that these structures identify preconditions follows directly from the fact that only level-preserving transformations are admitted and that comparisons are required to be natural and synchronous; this organizes which changes preserve the conditions under which responsible-AI claims could be transferred. We agree, however, that an explicit mapping of a non-deterministic retraining process onto admissible paths and the resulting quotient would make the alignment with continuity clearer. We will add a short illustrative example in the revised manuscript. revision: yes

  2. Referee: [Abstract] Abstract, strong interpretation paragraph: the requirement of 'mutual trustworthiness-preserving reachability, expressed through state isomorphism or natural isomorphism of realized histories' is presented as delivering the hierarchy, but without an explicit example of an admissible lifecycle path or a time-synchronous natural transformation, it is impossible to verify whether the isomorphism condition captures or abstracts away environment-dependent trustworthiness shifts that would block claim transfer in practice.

    Authors: The strong interpretation is defined in the manuscript as mutual reachability via isomorphisms that preserve the trustworthiness-level function. The abstract does not exhibit a concrete path or natural transformation, but the definitions of admissible lifecycle paths (trustworthiness-level-preserving morphisms) and time-synchronous natural transformations are given explicitly in Sections 3 and 4. We accept that an example would allow verification against environment-dependent shifts. We will therefore include, in the revision, a brief worked example of two histories related by a time-synchronous natural transformation, indicating when an environmental change blocks or permits the isomorphism. revision: yes

Circularity Check

0 steps flagged

No circularity: standard category-theoretic construction applied to defined AI concepts

full rationale

The paper defines an AI system type via a datum (techno-function, trustworthiness profile, trustworthiness-level function), then builds reachability categories from admissible paths, histories as functors, and comparisons as natural transformations. These are standard CT primitives applied to explicitly introduced objects; the weak/strong identity interpretations are direct readings of equality or isomorphism within that structure. No fitted parameters, no self-citation chains, no renaming of empirical patterns, and no reduction of the claimed preconditions to the inputs by construction. The framework is self-contained as a formal modeling exercise resting on ordinary category theory.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 5 invented entities

The framework rests on standard category theory together with several newly defined constructs that have no independent empirical support outside the formalization itself.

axioms (1)
  • standard math Standard axioms of category theory (existence of categories, functors, and natural transformations)
    The entire model is constructed inside this mathematical setting.
invented entities (5)
  • AI system type datum consisting of techno-function, trustworthiness profile, and trustworthiness-level function no independent evidence
    purpose: To specify the type of an AI system for identity analysis
    Introduced as the basic object of the formalization.
  • admissible lifecycle paths no independent evidence
    purpose: To connect profile-relative states under trustworthiness-preserving transformations
    Defined to restrict the allowed changes between states.
  • reachability category no independent evidence
    purpose: Quotient of admissible paths to obtain a category of reachable states
    New categorical structure obtained after quotienting.
  • temporally admissible functors no independent evidence
    purpose: To represent AI system histories
    Introduced to model the temporal organization of persistence.
  • time-synchronous natural transformations no independent evidence
    purpose: To compare realized histories
    Defined for synchronic comparisons between histories.

pith-pipeline@v0.9.1-grok · 5786 in / 1593 out tokens · 42874 ms · 2026-07-02T00:24:20.069163+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

137 extracted references · 3 canonical work pages · 1 internal anchor

  1. [1]

    2016 , howpublished =

    2016

  2. [2]

    2019 , month = apr, institution =

    Ethics guidelines for trustworthy. 2019 , month = apr, institution =

    2019

  3. [3]

    Recommendation on the Ethics of Artificial Intelligence , year =

  4. [4]

    Conformity assessment under the

    Thelisson, Eva and Verma, Himanshu , journal=. Conformity assessment under the. 2024 , publisher=

    2024

  5. [5]

    Conformity assessments and post-market monitoring:

    M. Conformity assessments and post-market monitoring:. Minds and Machines , volume=. 2022 , publisher=

    2022

  6. [6]

    Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency , pages=

    Nolte, Henrik and Rateike, Miriam and Finck, Mich. Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency , pages=

    2025

  7. [7]

    2019 , organization=

    Towards an ethical code for data-based business , author=. 2019 , organization=

    2019

  8. [8]

    The law of recitals in European Community legislation , author=. ILSA J. Int'l & Comp. L. , volume=. 2008 , publisher=

    2008

  9. [9]

    Trustworthy artificial intelligence and the

    Laux, Johann and Wachter, Sandra and Mittelstadt, Brent , journal=. Trustworthy artificial intelligence and the. 2024 , publisher=

    2024

  10. [10]

    Proceedings of the Conference on Fairness, Accountability, and Transparency (FAT* '19) , year =

    Margaret Mitchell and Simone Wu and Andrew Zaldivar and Parker Barnes and Lucy Vasserman and Ben Hutchinson and Elena Spitzer and Inioluwa Deborah Raji and Timnit Gebru , title =. Proceedings of the Conference on Fairness, Accountability, and Transparency (FAT* '19) , year =

  11. [11]

    Regulating the undefined: Addressing systemic risks in the Digital Services Act (with an appendix on the

    Loi, Michele and Fabbri, Matteo and Ferrario, Andrea , journal=. Regulating the undefined: Addressing systemic risks in the Digital Services Act (with an appendix on the. 2025 , publisher=

    2025

  12. [12]

    Nature , volume=

    Role play with large language models , author=. Nature , volume=. 2023 , publisher=

    2023

  13. [13]

    Communications of the ACM , volume=

    Talking about large language models , author=. Communications of the ACM , volume=. 2024 , publisher=

    2024

  14. [14]

    2024 , month = jun, institution =

    2024

  15. [15]

    2022 , url =

    European declaration on. 2022 , url =

    2022

  16. [16]

    2001 , month = nov, howpublished =

    2001

  17. [17]

    Mitigating the adverse effects of

    W. Mitigating the adverse effects of. Global Business and Organizational Excellence , volume=. 2024 , publisher=

    2024

  18. [18]

    2020 , month = feb, url =

    2020

  19. [19]

    2017 , month = apr, url =

    2017

  20. [20]

    2024 , publisher=

    Anthropomorphising machines and computerising minds: The crosswiring of languages between Artificial Intelligence and Brain & Cognitive Sciences , author=. 2024 , publisher=

    2024

  21. [21]

    Quine, W. V. O. , title =. 1969 , address =

    1969

  22. [22]

    2001 , address =

    Wiggins, David , title =. 2001 , address =

    2001

  23. [23]

    Lowe, E. J. , title =. Artefact Kinds: Ontology and the Human-Made World , editor =. 2014 , pages =

    2014

  24. [24]

    , title =

    Pinter, Charles C. , title =. 2014 , publisher =

    2014

  25. [25]

    , title =

    Olson, Eric T. , title =. 2024 , publisher =

    2024

  26. [26]

    2020 , publisher=

    In AI we trust: Ethics, artificial intelligence, and reliability , author=. 2020 , publisher=

    2020

  27. [27]

    Akbulut, Canfer and Weidinger, Laura and Manzini, Arianna and Gabriel, Iason and Rieser, Verena , booktitle=

  28. [28]

    Formalizing trust in artificial intelligence:

    Jacovi, Alon and Marasovi. Formalizing trust in artificial intelligence:

  29. [29]

    2020 , address =

    Bradford, Anu , title =. 2020 , address =

    2020

  30. [30]

    2020 , address =

    Petit, Nicolas , title =. 2020 , address =

    2020

  31. [31]

    Proposed

    Kazim, Emre and G. Proposed. AI and Ethics , volume=. 2023 , publisher=

    2023

  32. [32]

    Floridi, Luciano , journal=. The. 2021 , publisher=

    2021

  33. [33]

    Ho, Calvin Wai-Loon and Caals, Karel , journal=. How the. 2024 , publisher=

    2024

  34. [34]

    W. The. AI and Ethics , volume=. 2025 , publisher=

    2025

  35. [35]

    Institutionalised distrust and human oversight of artificial intelligence: towards a democratic design of

    Laux, Johann , journal=. Institutionalised distrust and human oversight of artificial intelligence: towards a democratic design of. 2024 , publisher=

    2024

  36. [36]

    A blueprint for auditing generative

    M. A blueprint for auditing generative. 2025 , publisher=

    2025

  37. [37]

    Auditing large language models:

    M. Auditing large language models:. 2024 , publisher=

    2024

  38. [38]

    Smuha, Nathalie A and Rengers, Emma and Harkens, Adam and Li, Wenlong and MacLaren, James and Piselli, Riccardo and Yeung, Karen , journal =. How the. 2021 , month = aug, doi =

    2021

  39. [39]

    Twenty-four years of empirical research on trust in

    Benk, Michaela and Kerstan, Sophie and von Wangenheim, Florian and Ferrario, Andrea , journal=. Twenty-four years of empirical research on trust in. 2025 , publisher=

    2025

  40. [40]

    An overview of key trustworthiness attributes and

    Mattioli, Juliette and Sohier, Henri and Delaborde, Agn. An overview of key trustworthiness attributes and. AI and Ethics , volume=. 2024 , publisher=

    2024

  41. [41]

    Trustworthy artificial intelligence:

    Kaur, Davinder and Uslu, Suleyman and Rittichier, Kaley J and Durresi, Arjan , journal=. Trustworthy artificial intelligence:. 2022 , publisher=

    2022

  42. [42]

    Ala-Pietil. The. 2020 , publisher=

    2020

  43. [43]

    Z-Inspection

    Zicari, Roberto V and Brodersen, John and Brusseau, James and D. Z-Inspection. IEEE Transactions on Technology and Society , volume=. 2021 , publisher=

    2021

  44. [44]

    Justifying our credences in the trustworthiness of AI systems:

    Ferrario, Andrea , journal=. Justifying our credences in the trustworthiness of AI systems:. 2024 , publisher=

    2024

  45. [45]

    Trustworthy

    Li, Bo and Qi, Peng and Liu, Bo and Di, Shuai and Liu, Jingen and Pei, Jiquan and Yi, Jinfeng and Zhou, Bowen , journal=. Trustworthy. 2023 , publisher=

    2023

  46. [46]

    2019 , howpublished =

    Ethics guidelines for. 2019 , howpublished =

    2019

  47. [47]

    and Davis, James H

    Mayer, Roger C. and Davis, James H. and Schoorman, F. David , title =. Academy of Management Review , volume =. 1995 , doi =

    1995

  48. [48]

    Ethics-based auditing to develop trustworthy

    M. Ethics-based auditing to develop trustworthy. Minds and Machines , volume=. 2021 , publisher=

    2021

  49. [49]

    Trust, distrust and commitment , author=. No. 2014 , publisher=

    2014

  50. [50]

    Synthese , volume=

    The fine-grained metaphysics of artifactual and biological functional kinds , author=. Synthese , volume=. 2009 , publisher=

    2009

  51. [51]

    A Trustworthiness-based metaphysics of

    Ferrario, Andrea , booktitle=. A Trustworthiness-based metaphysics of

  52. [52]

    Philosophical Explorations , volume=

    The ontology of artifacts , author=. Philosophical Explorations , volume=. 2004 , publisher=

    2004

  53. [53]

    , year =

    Gallois, A. , year =

  54. [54]

    Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society , volume=

    Social misattributions in conversations with large language models , author=. Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society , volume=

  55. [55]

    Ascribing functions to technical artefacts:

    Vermaas, Pieter E and Houkes, Wybo , journal=. Ascribing functions to technical artefacts:. 2003 , publisher=

    2003

  56. [56]

    Journal of Medical Ethics , volume=

    Are clinicians ethically obligated to disclose their use of medical machine learning systems to patients? , author=. Journal of Medical Ethics , volume=. 2025 , publisher=

    2025

  57. [57]

    A moving target in

    Hatherley, Joshua , journal=. A moving target in. 2025 , publisher=

    2025

  58. [58]

    Diachronic and synchronic variation in the performance of adaptive machine learning systems:

    Hatherley, Joshua and Sparrow, Robert , journal=. Diachronic and synchronic variation in the performance of adaptive machine learning systems:. 2023 , publisher=

    2023

  59. [59]

    2019 , publisher=

    Transparency in algorithmic and human decision-making: Is there a double standard? , author=. 2019 , publisher=

    2019

  60. [60]

    AI & society , volume=

    Artificial intelligence in medicine and the disclosure of risks , author=. AI & society , volume=. 2021 , publisher=

    2021

  61. [61]

    Machine learning applications in healthcare and the role of informed consent:

    Lorenzini, Giorgia and Shaw, David Martin and Arbelaez Ossa, Laura and Elger, Bernice Simone , journal=. Machine learning applications in healthcare and the role of informed consent:. 2023 , publisher=

    2023

  62. [62]

    Diagnosing diabetic retinopathy with artificial intelligence:

    Ursin, Frank and Timmermann, Cristian and Orzechowski, Marcin and Steger, Florian , journal=. Diagnosing diabetic retinopathy with artificial intelligence:. 2021 , publisher=

    2021

  63. [63]

    Towards a taxonomy for the opacity of

    Facchini, Alessandro and Termine, Alberto , booktitle=. Towards a taxonomy for the opacity of. 2021 , organization=

    2021

  64. [64]

    How the machine `thinks':

    Burrell, Jenna , journal=. How the machine `thinks':. 2016 , publisher=

    2016

  65. [65]

    Journal of Medical Internet Research , volume=

    Algorithm change protocols in the regulation of adaptive machine learning--based medical devices , author=. Journal of Medical Internet Research , volume=. 2021 , publisher=

    2021

  66. [66]

    Machine learning operations (

    Kreuzberger, Dominik and K. Machine learning operations (. IEEE Access , volume=. 2023 , publisher=

    2023

  67. [67]

    Journal of Systems and Software , volume=

    Adaptive data quality scoring operations framework using drift-aware mechanism for industrial applications , author=. Journal of Systems and Software , volume=. 2024 , publisher=

    2024

  68. [68]

    Neural Networks , volume=

    Change-point detection in time-series data by relative density-ratio estimation , author=. Neural Networks , volume=. 2013 , publisher=

    2013

  69. [69]

    IEEE Transactions on Knowledge and Data Engineering , volume=

    Concept drift detection delay index , author=. IEEE Transactions on Knowledge and Data Engineering , volume=. 2022 , publisher=

    2022

  70. [70]

    From concept drift to model degradation:

    Bayram, Firas and Ahmed, Bestoun S and Kassler, Andreas , journal=. From concept drift to model degradation:. 2022 , publisher=

    2022

  71. [71]

    Learning under concept drift:

    Lu, Jie and Liu, Anjin and Dong, Fan and Gu, Feng and Gama, Joao and Zhang, Guangquan , journal=. Learning under concept drift:. 2018 , publisher=

    2018

  72. [72]

    IEEE Access , volume=

    The robustness of counterfactual explanations over time , author=. IEEE Access , volume=. 2022 , publisher=

    2022

  73. [73]

    ACM Computing Surveys (CSUR) , volume=

    A survey of methods for explaining black box models , author=. ACM Computing Surveys (CSUR) , volume=. 2018 , publisher=

    2018

  74. [74]

    A multivocal review of

    Eken, Beyza and Pallewatta, Samodha and Tran, Nguyen and Tosun, Ayse and Babar, Muhammad Ali , journal=. A multivocal review of. 2025 , publisher=

    2025

  75. [75]

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

    The hidden assumptions behind counterfactual explanations and principal reasons , author=. Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency , pages=

    2020

  76. [76]

    Proceedings of the 3rd ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization , pages=

    Setting the right expectations: Algorithmic recourse over time , author=. Proceedings of the 3rd ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization , pages=

  77. [77]

    2015 , school=

    Accountable algorithms , author=. 2015 , school=

    2015

  78. [78]

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

    The philosophical basis of algorithmic recourse , author=. Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency , pages=

    2020

  79. [79]

    Canadian Journal of Philosophy , volume=

    On the opacity of deep neural networks , author=. Canadian Journal of Philosophy , volume=. 2023 , publisher=

    2023

  80. [80]

    2015 , publisher=

    More kinds of being: A further study of individuation, identity, and the logic of sortal terms , author=. 2015 , publisher=

    2015

Showing first 80 references.